Upload
others
View
4
Download
0
Embed Size (px)
Citation preview
1
Electrooptic Logic AND Gate Using a
Single Microring Resonator
M. Rakib Uddin*, Hafizuddin Helmi, Nur’azmina Lingas and Zainidi Haji Abdul Hamid
Electrical and Electronic Engineering Programme Area, Faculty of Engineering
Universiti Teknologi Brunei (UTB), Gadong, Brunei Darussalam
Abstract - In this paper, an electrooptic logic AND gate is
demonstrated using a single microring resonator. The
main principle of the logic operation is the resonant shift
due to the electro-optic effect which is responsible to
change the effective index of the device material. The
logic operation is justified by simulation with gate output
in the domain of optical spectrum as well as timing
diagrams. Extinction ratio of about 30 dB was recorded
from the gate output spectrum. Timing diagrams test
was performed with a 10 Gbps digital input signal and a
clear AND output was achieved.
Keywords- Silicon photonics; micro-ring resonator;
Photonic AND gate.
I. INTRODUCTION
Photonic technology in computing is important for
future high speed computers and communications [1], [2]. In optical signal processing, like other photonic devices and circuits, optical logic gates have also received considerable attention [3]-[5]. In the field of photonic computer and communications, the logic units are the main building blocks and in communication networks they can enable many advanced functions such as all-optical bit-pattern recognition, all-optical packet header and payload separation, all-optical bit error rate monitoring, all-optical label swapping, all-optical packet drop and so on. Microring resonator has potential applications in photonic computing. In references [6], [7] electro-optic logic gates are demonstrated using microring resonator; however, they used two cascaded rings to achieve the basic logic functions. In this paper, we propose a quite simple circuit using a single mirroring resonator instead of two cascaded rings to get logic AND logic functions. We design, simulate and demonstrate the results of logic AND gate. The gate is designed using a single micro ring resonator. The gate operation is verified by the optical spectrums as well as by the waveforms of 10 Gb/s data inputs.
II. PRINCIPLE OF OPERATION
A micro-ring resonator with external voltage
applied exhibits a property of resonant shift [8], [9]. These changes can be achieved by various methods and amongst all methods, the convenient and instant method is the electric fields passing through the
materials which in turn causes an electro-optic effect, i.e., the material index changes and then the resonance is shifted. Using this behaviour, logic AND gate function is demonstrated. The schematic of the logic circuit using a single ring along with the symbol and the truth table of AND gate is shown in Fig. 1.
Optical Input Optical Gate Output
Gate Inputs (Electrical)
A B
A B Output
0 0 0
0 1 0
1 0 0
1 1 1
A
B Output
(a)
(b) (c)
Fig. 1 The schematic of the logic circuit using a single ring along with the symbol and the truth table of AND gate.
Fig. 1 (a) is the schematic diagram of the single bus waveguide ring resonator. The single bus waveguide microring resonator has only one straight waveguide which has two ports including input port and throughput port. The straight waveguide is closely located with the ring waveguide. The diameter of the ring is dependent on the requirements of the device. The gap between the straight waveguide and the ring is also an important parameter. The coupling coefficient is a function of the gap. The waveguide dimensions are also depends on the design requirements. The typical width of the waveguide is 500 nm and the height of the waveguide is 220 nm. The gap is about 100 nm. An electrical excitation is applied to the ring resonator which is shown by “A, B” in the schematic diagram in Fig. 1 (a). The symbolic diagram of the AND gate is shown in Fig. 1 (b). The truth table of the logic AND functions are shown in Fig. 1 (c). The definition of AND logic is
2
the output is logic HIGH only if the inputs are logic HIGH, otherwise the outputs are logic LOW which is shown in the truth table in Fig. 1 (c).
The optical micro ring resonator used for the demonstration in Fig. 1 has a frequency of resonance at λr and with external voltage the resonant shift to λr’. The gate function is based on the resonant shift. The phenomena of the resonant shift is shown in Fig. 2 schematically for different gate voltages of AND operations. When an optical CW light is sent to the ring resonator input port with the wavelength of λr, it gives an optical throughput output with low power, which can be classified as optical logic ‘0’. When sufficient electrical supply is sent to the resonator input port, the ring resonator output optical power increases to the point it can be considered as logic ‘1’.
In Fig. 2 (a), it is shown schematically that for the voltage inputs 0, 0; the resonant shift is not sufficient; So that the optical output is low and is it considered as logic “0”. In Fig. 2 (b), it is shown that for the voltage inputs either 0, 1; or 1, 0; the resonant shift is not sufficient too; So that the optical output is low and is it considered as logic “0”. Whereas when both the electrical inputs are high (1, 1), the resonant shift is sufficient which is shown in Fig. 2 (c). So that the optical output is high enough which can be considered as logic “1”. This is the schematic of logic AND gate operation. In the figure (Fig. 2), the wavelengths λr and λr’ represents the original and the shifted wavelength, respectively.
Fig. 2 The schematic of optical spectrum for logic AND function.
Fig. 2 (a) shows the low logic output when both the electrical logic inputs are low, i.e., both are 0s. In this case, both the light wavelengths are located at the same position which means the measured light intensity at the original wavelength and the wavelength due to two electrical inputs A, B as LOW are same as logic LOW. Fig. 2 (b) shows the schematic of logic low output when either inputs is 0 or 1. If only one input is high, it is not enough to shift the resonance enough to get a higher optical power at the output. So, the output is logic low. In this case both the original resonant wavelength and the wavelength due the either A or B logic HIGH are located closely so that the intensity of the light at throughput port is still logic LOW. Fig. 2 (c) shows the 1, 1 operation. When both the electrical inputs are high, the resonance is shifted enough so that the output is high enough to consider the logic high, ‘1’. In this case it is shown that the original resonant wavelength and the wavelength due to the excited A, B are located at different location which makes sure that the light intensity at the original resonant point is high enough compare to the other three cases. This high intensity is
considered as logic HIGH. In all three figures in Fig. 2 the two dotted lines represent the two logic inputs. Base on this principle, we design our logic device in the simulation environment using a commercially certified software called “Lumerical (www.lumerical.com). In the software, there are four modules. We used the module, “INTERCONNECT” to design and simulated the logic operation of the AND gate based on a single photonic microring resonator. The justification of the logic operation in spectrum domain as well as using timing diagrams are conducted using this software which will be described in the following sections with output results. The design of light source, the ring resonator, electrical signals and the optical spectrum analyser were performed using the same “INTERCONNECT” environment.
3
III. LOGIC SIMULATION AND RESULTS
Fig. 3 shows the photonic AND gate input-output
with optical spectrum. A commercially certified simulation software called “Lumerical Solutions” [https://www.lumerical.com/] was used to simulate the logic function. The spectrum results in Fig. 3 show
that the optical intensity is LOGIC HIGH only when both the inputs are LOGIC HIGH which is the truth of AND gate. For four conditions (00, 01, 10, 11), the output intensities are -18 dB, -23 dB, -18 dB and 10 dB. 10 dB is logic high whereas -18 or lower dBs are logic low. So, output is LOGIC HIGH when
-120
-100
-80
-60
-40
-20
0
20
1547 1548 1549 1550 1551 1552 1553
Outp
ut
Po
wer
(d
Bm
)
Wavelength (nm)
0,0
-120
-100
-80
-60
-40
-20
0
20
1547 1548 1549 1550 1551 1552 1553
Outp
ut
Po
wer
(d
Bm
)
Wavelength (nm)
0,1
-120
-100
-80
-60
-40
-20
0
20
1547 1548 1549 1550 1551 1552 1553
Outp
ut
Po
wer
(d
Bm
)
Wavelength (nm)
1,0
-120
-100
-80
-60
-40
-20
0
20
1547 1548 1549 1550 1551 1552 1553
Outp
ut
Po
wer
(d
Bm
)
Wavelength (nm)
1,1
≈ -23dBm
≈ 10dBm
≈ -18dBm
≈ -18dBm
Fig. 3 Simulation results: AND logic output in spectrum domain.
4
Fig. 4 Simulation results: AND logic output timing
diagrams.
both inputs are high which is the AND gate truth
function. The spectrum results in Fig. 3 were
measured by the simulation and recorded using the
optical spectrum analyser. The input light was set to a
single wavelength light source to fit a certain resonant
wavelength. Fig 4 shows the timing diagrams of the
AND gate outputs simulated by the software. The
AND gate output was recorded using the through port
of the ring resonator. The timing diagrams also
verifies the AND operation. In Fig. 4, “A” and “B” are
the input waveforms and “AND” is the output. In the
figure it is shown that when inputs are 0, 0, the output
is “0. When inputs are 1, 0, or 0, 1, the output is
also“0”. When both inputs are 1, 1, the output is “1”
which are the truth functions of AND gate.
IV. CONCLUSION
In this paper, we have simulated and demonstrated a
novel scheme of digital micro-photonic logic gate
based on the opto-electronic effect in silicon photonic
micro ring resonator. The AND logic functions were
verified by both the optical spectrum and digital data
signal at the rate of 10 Gbps. Both the optical
spectrum and the digital waveforms, proved the AND
logic operation using a single micro ring resonator. It
is noted that the optical micro-ring resonator can be a
promising micro device for photonic logic design.
Since the AND gate is the basic logic gate, it can be
the basis for the micro-photonic digital building
blocks for future optical computers and
communication applications.
ACKNOWLEDGEMENT
The authors gratefully acknowledge use of Graduate
Studies and Research Office Grant [UTB/GSR/1/2016
(1)] at Universiti Teknologi Brunei (UTB), Brunei
Darussalam.
REFERENCES
[1] H. J. Caulfield and S. Dolev, “Why future supercomputing requires optics,” Nat. Photon., vol. 4, pp. 261–263, 2010. [3] D. A.
B. Miller, “The role of optics in computing,” Nat. Photon., vol. 4, p.
406, 2010. [2] D. A. B. Miller, “The role of optics in computing,” Nat. Photon.,
vol. 4, p. 406, 2010.
[3] H. Yoo H. J. lee, Y. D. Jeong, and Y. H. Won, “All-optical logic gates using absorption modulation of an injection-locked Fabry-
Perot laser diode,” in proc. Photonics in Switching, Greece, Oct. 16-
18, 2006. [4] M. R. Uddin, J. S. Lim, Y. D. Jeong and Y. H. Won, "All-optical
digital logic gates using single-mode Fabry-Perot laser diodes,"
Photonics Technology Letters, IEEE, vol. 21, no. 19, 2010. [5] G. Berrettini, A. Simi, A. Malacarne, A. Bogoni, and L. Poti,
“Ultrafast integrable and reconfigurable XNOR, AND, NOR, and
NOT photonic logic gate,” IEEE Photon. Technol. Lett., vol. 18, no. 8, 2006.
[6] Y. Tian et all, “Simulation and Demonstration of Directed
XOR/XNOR Logic Gates Using Two Cascaded Microring Resonators”, IEEE Photon. Journal, vol. 8, no. 2, 2016.
[7] Y. Tian et al., “Electro-optic directed AND/NAND logic circuit
based on two parallel microring resonators”, Opt. Express, vol.20, no. 15, 2012.
[8] M. Elshoff and O. Rauntenberg, "Design and Modelling of Ring
Resonators used as Optical Filters for Communications Applications,"
Universidad Publica de Navarra, Pamplona, Spain, 2010.
[9] M. Rakib Uddin, and Y. H. Won, "Rib Waveguide-based Athermal Micro-ring Resonator", Optical and Quantum Electronics, Vol. 47, No. 8,
2015.
5
Design of a Half-bridge Synchronous Rectifier with
Current Doubler Resonant Converter for Optimizing
Solar Powered Surveillance System
Ryann Alimuin1, Elmer Dadios2, Aldrich Guiron1, Ramon Asio1, Paul Joshua Miranda1, Razelle R. Vale1 and
Christian John Fernandez1 1Technological Institute of the Philippines – Quezon City
2De La Salle University - Manila
[email protected], [email protected]
Abstract — the power optimizer utilizes a resonant converter
which is integrated in a surveillance system. The resonant
converter has a low input voltages boost (increase) and high input
voltages buck (decrease) characteristic for a target specification of
12V. The power optimizer has also a bidirectional property where
it allows the charging of the battery while supplying the load.
Power optimization was achieved through synchronous
rectification in which the losses were significantly reduced since
MOSFETs are used for fast switching. The results show that the
peak efficiency is at nominal at 12V. The efficiency of the power
optimizer was evaluated with varying loads. For an 8V supply
input, the efficiency ranges from 56% to 80%, at 12V input, it
ranges from 54% to 98% and at 17V input ranges from 45% to
98% with load variation from 1Ω to 1000 Ω.
Key Terms: Current Doubler, Load Variation, MOSFET, Power
optimization, Resonant, Resonant converter
I. INTRODUCTION
Power supply has been doing a vital role in each and every
machinery and electronic equipment because it powers up the
whole system. Studies show that using the right and the best
power supply for the system is most important while using the
wrong power supply in the system can lead to major problems
like shortening the lifespan of the system, and worst case, can
destroy the whole system. The most commonly used power
supply nowadays is the Switch Mode Power Supply (SMPS). It
was invented to overcome the disadvantages of the linear power
supply such as large power loss due to the use of large electronic
equipment. Linear Mode Power Supply can only be used as a
step-down regulator, unlike the SMPS, it can step-up and step-
down the voltage. It can convert the input voltage into higher
one or into lower one. SMPS are used especially to have high
supply efficiency for high power application, and to maintain
the lifespan of the battery and its external temperature.
However, the problem in using SMPS which uses Pulse Width
Modulation (PWM) is its hard switching characteristic that is a
major cause of the switching loss [1].
Figure 1 shows the difference in losses between hard
switching and soft or resonant switching.
Figure 1. Current and Voltage waveforms of Hard and
Resonant Switching [2]
Resonant converters are DC-TO-DC designed to overcome the
disadvantages of conventional converters. To achieve high
power density, converters must be designed with high
frequency. Conventional converters can achieve high efficiency
compared to linear but the problem is when the switching
frequency increases also the losses increases due to switching
frequency. Resonant converters can overcome the
disadvantages of the conventional converters because of its
ability to achieve high efficiency by soft-switching. It is
classified into three types Conventional Converter, Quasi-
resonant Converter and Multiresonant Converter (shown in
figure 2).
Figure 2. Classification of Resonant Converter [3]
Resonant converter can overcome the problems with the other
SMPS topologies. Using the soft-switching technique it can
6
reduce losses and can generate EMI better performance. With
the use of synchronous rectification, the efficiency of the typical
resonant converter can be raised more. Synchronous
rectification with current doubler is used to reduce switching
and conducting losses of active switches. Converter operating
at resonant mode can have improved efficiency because the
impedance between input and output are at its minimum [3].
II. FUNCTIONAL DIAGRAM OF THE POWER OPTIMIZER
UTILIZING RESONANT CONVERTER
The power optimizer includes the switching, the resonant
converter circuit, and the rectifier / filter. Soft switching
technique will be used which reduces the possibility of having
very high loss in electronic switches. It is usually a state of Zero
Voltage Switches (ZVS) or Zero Current Switches (ZCS) that
increases the conversion efficiency within power electronics
topologies [3]. The resonant converter circuit uses equal
capacitance and inductance to enhance the wave formation of
either the current or the voltage through the switching
component thus, no power dissipation when switching happens
due to ZVS or ZCS. Rectifier and filter part are used to
smoothen and to have a steady or stable DC voltage output.
Once the power optimizer is placed, the energy harvested from
the solar panel to the battery will be maximized and achieved a
much higher efficiency [4].
Figure 3. Functional Diagram of the Power optimizer
The functional diagram of the power optimizer is shown in
figure 3. First, the input of the converter which is from the solar
panel will be monitored by the voltage sense of the MOSFET
DRIVER. If the sensed voltage is lower than the intended output
voltage, the converter will perform boost mode, while buck
mode will be perform when the output voltage is higher than the
intended output voltage. The buck and boost mode has a
characteristic of synchronous switching which will help to
make a resonant conversion [6]. Regulated output will be done
by filtering the voltage output to lessen the voltage ripple, the
current doubler placed at the end of the converter is used to
provide a lesser voltage ripple [4]. The power optimizer also has
a feature of bidirectional switching [3]. When the output that
from the converter is enough, it will simultaneously supply the
battery and the camera. The bidirectional circuit is also
responsible for monitoring the output voltage of both converter
and battery, and also to ensure that the power that would be
powering up the load will be coming from either of the two that
which has higher voltage output.
III. DESIGN MODELLING OF THE POWER OPTIMIZER
The following equations are used to solve for the parameters of
the resonant converter to provide an efficient power optimizer.
Output power based from target efficiency
PO in Watts
𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 (%) =𝑃𝑜
𝑃𝑖
The value of the output voltage and current based from the
target efficiency must be generalized.
VO = in Voltage
IO = in Ampere
Soft-start Capacitor
When the RUN pin voltage is higher than 1.5V, an internal
1.2µA current source charges soft-start capacitor CSS at the SS
pin. The ITH voltage is then clamped to the SS voltage while
CSS is slowly charged during start-up. This “soft-start”
clamping prevents abrupt current from being drawn from the
input power optimizer. [3]
𝑇𝐼𝑅𝑀𝑃 = 𝑅𝑈𝑁 𝑝𝑖𝑛
1.2 𝜇𝐴−𝑖𝑛𝑡𝑒𝑟𝑛𝑎𝑙 𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑠𝑜𝑢𝑟𝑐𝑒∗ 𝐶𝑆𝑆
Frequency Synchronization and Frequency Setup
The phase-locked loop allows the internal oscillator to be
synchronized to an internal source via PLLIN pin. To set the
frequency of the power optimizer, the PLLFLTR pin should
have an input voltage from 0 V to 2.4 V. [3]
Inductor Selection
𝐿𝐵𝑂𝑂𝑆𝑇 >𝑉𝐼𝑁(𝑀𝐼𝑁)
2 ⋅ (𝑉𝑂𝑈𝑇 − 𝑉𝐼𝑁(𝑀𝐼𝑁)) ⋅ 100
𝑓 ⋅ 𝐼𝑂𝑈𝑇(𝑀𝐴𝑋) ⋅ %𝑅𝑖𝑝𝑝𝑙𝑒 ⋅ 𝑉𝐼𝑁(𝑀𝐴𝑋)
7
𝐿𝐵𝑈𝐶𝐾 >𝑉𝑂𝑈𝑇 ⋅ (𝑉𝐼𝑁(𝑀𝐴𝑋) − 𝑉𝑂𝑈𝑇) ⋅ 100
𝑓 ⋅ 𝐼𝑂𝑈𝑇(𝑀𝐴𝑋) ⋅ %𝑅𝑖𝑝𝑝𝑙𝑒 ⋅ 𝑉𝐼𝑁(𝑀𝐴𝑋)
𝐑𝐒𝐄𝐍𝐒𝐄Selection and Maximum Output Current
𝐼𝑂𝑈𝑇(𝑀𝐴𝑋,𝐵𝑂𝑂𝑆𝑇) = (160𝑚𝑉
𝑅𝑆𝐸𝑁𝑆𝐸−
∆𝐼𝐿
2) ∙
𝑉𝐼𝑁(𝑀𝐼𝑁)
𝑉𝑂𝑈𝑇
𝐼𝑂𝑈𝑇(𝑀𝐴𝑋,𝐵𝑈𝐶𝐾) = (130𝑚𝑉
𝑅𝑆𝐸𝑁𝑆𝐸+
∆𝐼𝐿
2)
BOOST MODE:
𝑅𝑆𝐸𝑁𝑆𝐸(𝑀𝐴𝑋) =2 ∙ 160𝑚𝑉 ⋅ 𝑉𝐼𝑁(𝑀𝐼𝑁)
2 ⋅ 𝐼𝑂𝑈𝑇(𝑀𝐴𝑋,𝐵𝑂𝑂𝑆𝑇) ⋅ 𝑉𝑂𝑈𝑇 + ∆𝐼𝐿𝐵𝑂𝑂𝑆𝑇 ⋅ 𝑉𝐼𝑁(𝑀𝐼𝑁)
BUCK MODE
𝑅𝑆𝐸𝑁𝑆𝐸(𝑀𝐴𝑋) =2 ∙ 130𝑚𝑉
2 ⋅ 𝐼𝑂𝑈𝑇(𝑀𝐴𝑋,𝐵𝑈𝐶𝐾) − ∆𝐼𝐿𝐵𝑂𝑂𝑆𝑇
𝑪𝑰𝑵 and 𝑪𝑶𝑼𝑻Selection
𝐼𝑅𝑀𝑆 ≈ 𝐼𝑂𝑈𝑇(𝑀𝐴𝑋) ⋅𝑉𝑂𝑈𝑇
𝑉𝐼𝑁⋅ √
𝑉𝐼𝑁
𝑉𝑂𝑈𝑇− 1
NOTE: This formula has a maximum at 𝑽𝑰𝑵 = 2𝑽𝑶𝑼𝑻, where
𝑰𝑹𝑴𝑺 = 𝑰𝑶𝑼𝑻(𝑴𝑨𝑿).
𝑅𝑖𝑝𝑝𝑙𝑒(𝐵𝑜𝑜𝑠𝑡, 𝐶𝑎𝑝) =𝐼𝑂𝑈𝑇(𝑀𝐴𝑋) ⋅ (𝑉𝑂𝑈𝑇 − 𝑉𝐼𝑁(𝑀𝐼𝑁))
𝐶𝑂𝑈𝑇 ⋅ 𝑉𝑂𝑈𝑇 ⋅ 𝑓
𝑅𝑖𝑝𝑝𝑙𝑒(𝐵𝑢𝑐𝑘, 𝐶𝑎𝑝) =𝐼𝑂𝑈𝑇(𝑀𝐴𝑋) ⋅ (𝑉𝐼𝑁(𝑀𝐴𝑋) − 𝑉𝑂𝑈𝑇)
𝐶𝑂𝑈𝑇 ⋅ 𝑉𝐼𝑁(𝑀𝐴𝑋) ⋅ 𝑓
The steady ripple due to the voltage drop across the
ESR is given by:
∆𝑉𝐵𝑂𝑂𝑆𝑇,𝐸𝑆𝑅 = 𝐼𝐿(𝑀𝐴𝑋,𝐵𝑂𝑂𝑆𝑇) ⋅ 𝐸𝑆𝑅
∆𝑉𝐵𝑈𝐶𝐾,𝐸𝑆𝑅 = 𝐼𝐿(𝑀𝐴𝑋,𝐵𝑈𝐶𝐾) ⋅ 𝐸𝑆𝑅
Power MOSFET Selection and Efficiency
Considerations
𝑃𝐴,𝐵𝑂𝑂𝑆𝑇 = (𝑉𝑂𝑈𝑇
𝑉𝐼𝑁⋅ 𝐼𝑂𝑈𝑇(𝑀𝐴𝑋))
2
⋅ 𝜌𝑇
⋅ 𝑅𝐷𝑆(𝑂𝑁)
where ρT is a normalization factor (unity at 25°C) accounting
for the significant variation in on-resistance with temperature,
typically about 0.4%/°C as shown in Figure 9. For a maximum
junction temperature of 125°C, using a value ρT = 1.5 is
reasonable. [3]
𝑃𝐵,𝐵𝑈𝐶𝐾 =𝑉𝐼𝑁 − 𝑉𝑂𝑈𝑇
𝑉𝐼𝑁𝐼𝑂𝑈𝑇(𝑀𝐴𝑋)
2 ⋅ 𝜌𝑇
⋅ 𝑅𝐷𝑆(𝑂𝑁)
𝑃𝐶,𝐵𝑂𝑂𝑆𝑇 =(𝑉𝑂𝑈𝑇 − 𝑉𝐼𝑁)𝑉𝑂𝑈𝑇
𝑉𝐼𝑁2 𝐼𝑂𝑈𝑇(𝑀𝐴𝑋)
2 ⋅ 𝜌𝑇
⋅ 𝑅𝐷𝑆(𝑂𝑁) + 𝑘 ⋅ 𝑉𝑂𝑈𝑇2
⋅𝐼𝑂𝑈𝑇(𝑀𝐴𝑋)
𝑉𝐼𝑁⋅ 𝐶𝑅𝑆𝑆
Where CRSS is usually specified by the MOSFET
manufacturers. The constant k, which accounts for the loss
caused by reverse recovery current, is inversely proportional to
the gate drive current and has an empirical value of 1.7. [3]
𝑃𝐷,𝐵𝑂𝑂𝑆𝑇 =𝑉𝐼𝑁
𝑉𝑂𝑈𝑇⋅ (
𝑉𝑂𝑈𝑇
𝑉𝐼𝑁⋅ 𝐼𝑂𝑈𝑇(𝑀𝐴𝑋))
2
⋅ 𝜌𝑇 ⋅ 𝑅𝐷𝑆(𝑂𝑁)
𝑇𝐽 = 𝑇𝐴 + 𝑃 ⋅ 𝑅𝑇𝐻(𝐽𝐴)
INTVCC Regulator
Use of the EXTVCC input pin reduces the junction temperature
to:
𝑇𝐽 = 70∘𝐶 + 24𝑚𝐴 ⋅ 6𝑉 ⋅ 34∘𝐶 𝑊⁄ = 75∘𝐶
Output Voltage
𝑉𝑂𝑈𝑇 = 0.8𝑉 (1 +𝑅2
𝑅1)
8
IV. CIRCUIT SIMULATION
Integrated circuits are mostly used as the driver for the
circuit. An LTC3780 is used, it is a synchronous controller that
uses combined buck and boost circuit. Each stage is comprised
of two MOSFET switch. An active switch replaces the diode,
which is the MOSFET to make the converter more efficient. An
inductor separates the buck and the boost. This controller can
achieve 98% peak efficiency. It has phase-lockable frequency
of 200 kHz – 400 kHz. A fault protection comparator is and
foldback current limiter is provided inside the LTC3780. [3]
Figure 4. Schematic Diagram of the Circuit for Power
Optimizer
Figure 5. Voltage Sense Switching Circuit (Bidirectional
Circuit)
The bidirectional circuit is shown in Figure 5. It helps the power
optimizer to extend its capability in storing energy to the battery
while supplying the load.
Moreover, series of simulations and testing are needed to
determine if the target specifications are achieved.
Figure 6. Simulation output of the Design
Figure 7. Simulation of the Input and Output Power
The Figure 6 shows the target output power of the design.
Through the use of a power optimizer, the losses must be reduce
and the device must regulate at 12V. While Figure 7 shows the
simulation of the target input and output power of the power
optimizer to prove that the device optimizes the input source
from the solar panel.
The device should also test in various load to make sure
that the reliability is achieve. The power optimizer was tested
in different inputs which are the 12V, 8V, and 17V.
Figure 8. Power Optimizer Efficiency in Various Loads
Since it is used to optimize the harness energy from the sun, the
Figure below shows the output voltage gathered of the solar
panel at 10mins interval time.
9
Figure 9. Time vs Output Voltage of the Solar Panel
The Output power is also important in gathering data. The
figure below shows the time it takes for the device to have an
output power shown below.
Figure 10. Time and Output Power Analysis
The battery is also a part of the design because the power
optimizer has the bidirectional property that allows the
simultaneously charging the battery while supplying the load.
Figure 11. Battery Charging Time based on the Weather
Condition
V. SYSTEM APPLICATION
The power optimizer is to be integrated in a solar powered
surveillance system to prolong the time of consumption of the
CCTV camera. During daytime while the battery is charging it
is also supplying the load. At night since there is not enough
sunlight, the stored energy in the battery will now be used as the
source of energy that will supply to the load. The system is
comprised of a DC-DC converter used for power management.
Resonant converter is utilized because of its soft-switching
capability. The device optimizes the energy harnessed from the
sun and it helps to reduce the switching loss and conduction loss
encountered during conversion.
VI. CONCLUSION
The developed design is an effective tool that can be used for
power management. Multiple designs can be simulated to
increase performance in handling different load requirements.
The system can be integrated into any numerous load networks
and allowed to manage different inputs and output voltages.
VII. ACKNOWLEDGMENT
The authors would like to express their gratitude to the Engr.
Shearyl Arenas for the design guidance, Technological Institute
of the Philippines – Quezon City and De La Salle University –
Manila for the research collaboration and permission on the use
of relevant hardware and software in this activity.
VIII. REFERENCES:
[1] F. Z. X. X. D. J. a. Z. Q. Junming Zhang, "A Novel ZVS DC/DC Converter for High Power Application," IEEE Transactions for Power
Electronics, 2004, pp. 420-429.
[2] B. M., "Resonant Converter Topologies," STMicroelectronics, 1999.
[3] S.-P. Yang, J.-L. Lin and S.-J. Chen, "A Novel ZCZVT Forward
Converter With Synchronous Rectification," IEEE Transactions on
Power Electronics, vol. 21, pp. 912-922, 2006.
[4] A. Skinner, "Choosing the Right Topology," Boddo's Power Systems,
2009.
[5] G. Lakkas, "MOSFET Power Losses and How they Affect Power-Supply," Analog Applications Journal, pp. 22-26, 2016.
[6] K.-H. Liu, R. Oruganti and F. C. Y. Lee, "Quasi-Resonant Converter -
Topologies and Characteristics," IEEE Transactions on Power Electronics, vol. 1, pp. 62 - 71, 1987.
[7] J. S. Glaser and M. A. D. Rooij, "Current doubler Rectifier with Current
Ripple Cancellation," 2006.
[8] S.-M. (. Chen, Tsorng-Juu and K.-R. Hu, "Design, Analysis, and
Implementation of Solar Power Optimizer for DC Distribution System," IEEE, vol. 28, no. 4, pp. 1764-1772, 2013.
[9] L. Technology, "High Efficiency, Synchronous, 4-Switch Buck-Boost
Controller," Linear Technology Corporation, California, 2005.
[10] R. M. H., Power Electronics Handbook, 2001.
10
Electrical Energy from Self-Running
Magnetic Motor
Abdul Halim Ali, Ahmad Najmuddin Che Ismail Universiti Kuala Lumpur British Malaysian Institute
Kuala Lumpur, Malaysia
Abstract:
The use of magnetic motor to generate electricity ever since in
the 18 century. In most cases, external resources such as hydro
and wind are needed to power the magnetic motor before an
induce electricity can be produced. In this study, a magnetic
motor is design to self-rotate it’s rotor by naturally repulsion
and the attraction of magnetic field by arranging the
magnets into Halbach array. The self-rotation magnetic
motor sometime it is known as a machine that produce
“Free Electric Energy”. Based from the results the rotor is
rotating at the constant speed that produced the torque
that lead to the development of mechanical power.
Keywords: Induction Motor, Electrical Energy,
Halbach Array, Magnetic Motor.
I. INTRODUCTION
The term “free energy” is not maybe a gas station
giving away gas however this is not the case for Nikola
Tesla where he was the first one to identify “radiant
energy” where energy harvesting the Sun. Nikola Tesla is
the key researcher in free energy theories and invented
most of the free energy devices. Tesla introduced two free
energy theories. The earlier is known as Crooke’s
radiometer and later as “cosmic-ray motor” which he
claimed to be “thousands of times more powerful” as
compare to Crooke’s radiometer. Tesla’s free energy
concept was patented in 1901 as an “Apparatus for the
Utilization of Radiant Energy.” In 1932, Tesla claimed has
successful harnessed the cosmic rays. The radiant energy
receiver stored static electricity obtained from the air and
converted it to a usable form [1, 2]. However, Tesla’s free
energy are not from the magnetic motor generator that
produce the electricity.
In this study, a free energy is created from permanent
magnet motor without utilizes resources from outside such
as burning fossil fuels namely coal, petroleum and natural
gas [3] to induced voltage. The free energy comes from the
naturally repulsion and the attraction of magnetic field that
creates the motion of electric motors. This self-running
electric motors is attached to a turbine motor shaft which
resulting an induced voltage.
The term, "Free Energy" is widely used and
often abused in the industry. Many believe no such
thing of free energy, or whatsoever machine capable to
generate energy out of nothing. In others word, there are
no such things of “perpetual motion machine” that can do
work continues indefinitely without utilizing external.
II. MAGNETIC MOTOR
The first magnetic motor was first deployed in 1880 is
a direct current (DC) magnetic motor when direct current
was the only source of power, until Nikola Tesla invented
the alternative current (AC) magnetic motor in 1889 where
in 1886 the starting era of AC power system in the world.
The induction energy from AC magnetic motors need
external resources for operations such as hydro dams to
and windmills. With the exception of solar
power, 95% of the induction of electricity in the
world comes from electromagnet -based power
generating systems by setting magnets into
motion while wrapping windings of magnet wire
around the magnet to induce electricity. Since
than the research development of an AC magnetic
motor growth has never stop.
The configuration of the magnets in the self -
running magnetic motor in this study using
Halbach array. The simplest Halbach array
configuration as shown in figure 1 is creating
strong magnetic field at one side while cancelling
the field to near zero on the others side of the
array. The magnetic field lines of the Halbach
array is shown in figure 2.
Figure 1: Simple Halbach array configuration
Strongest side of magnetic field
Canceled side of magnetic field
11
Figure 2: The magnetic field lines of Halbach array [4]
The main advantages of using Halbach array where the
magnetic field strength produced is very strong and
increases the efficiency of the magnetic circuit as compare
with others arrays configurations. However, the main
drawback of such configuration it’s difficult to assemble
and the magnets are arranged in a direct or quasi-direct
repelling condition that will act to demagnetize their
neighboring magnets. The Halbach array configuration
major applications of the one-sided flux distribution
ranging from the simple refrigerator magnet to much
complicated application deployed in the Maglev train
(magnetic levitation).
The general description of the Halbach array
configuration magnetization pattern was given by [5], a
simple superposition of two trigonometric functions as
shown in equations 1 & 2
where λ denotes the wavelength and the magnetization
amplitude.
III. SELF-RUNNING MAGNETIC MOTOR DESIGN
The self-running magnetic motor is designed with 3 layers
as shown in figure 3. The basis of the design is based on
[6]. The most inner magnets consists of 10 magnets, the
middle magnets is made of 14 magnets and the outer
magnets is comprises of 21 magnets. The magnets field
arrangements follow Halbach array [7]. The middle and
the outer layers are the stator whereas the most-inner layer
act as rotor. The inner radius of the magnets is at 4.1cm,
the middle magnets radius at 6.4cm and outer magnets
radius at 1.7cm with an air-gap of 0.3cm between the
magnets. The whole radius of the design is at 12.5cm. The
magnets material use “neodymium (NdFeB) n52” is the
strongest permanent magnet commercially available in the
market. The property of the NdFeB n52 magnet is shown
in Table 1 and it is made from an alloy of neodymium, iron
and boron.
Table 1: Property of NdFeB n52 magnet Remanance (Br) Coersive
Force Hcb (Hc)
Intrensic
Coersive
Force Hcj (Hj)
Maximum
Energy Product
(BH) Max
mT G K A/M Oe K A/M Oe KJ/m³ MGOe
1430 14300 796 10000 876 11000 398 50
Figure 3: Self-running magnetic motor with
Halback magnetic field directions
Figure 4 shows the 3D design of the self-running
magnetic motor
Figure 4: The 3D design of self-running
magnetic motor
IV. SIMULATION RESULTS
The self-running magnetic motor design is tested using
Finite Elements simulations tool. The Finite Elements
FEMM4.2 is an open source magnetic motor that provide
wide range of possibilities to simulate the design. Figure 5
shows the magnetic field strength of the self-running
magnetic motor obtained from FEMM4.2
Figure 5: The magnetic field strength
12
Figure 6 shows the simulation results of relative
centrifugal force (RFC) produces by the self-running
magnetic motor. It is notice from the graph for 360 turn it of the self-running magnetic motor produces two cycles response. This is due that the rotor have ten magnets where the arrangement is set into two sets of Halbach array. Each set of Halbach array results in one complete cycle.
Figure 6: RFC waveform
To confirm that the rotor is rotating at a constant speed 10
cycles of 360 were simulated and the results are shown in
in Figure 7 and 8. Constant responses were recorded in
figure 7 showing that the rotor is rotating at a constant
speed.
Figure 7: RCF responses of 10 cycles
Figure 8 shows the harmonics of the 10 cycles. The
harmonics response of the 10 cycles shows the same as of
figure 6. This shows the design of the self-running
magnetic motor the rotor is rotating at the same frequency.
Torque is another important parameter that can be
measured from the simulation. Figure 9, shows the torque
response as expected it’s produced two cycles from 360
turn from the self-running magnetic motor as can be seen in figure 6 of RCF.
Figure 8: RCF Harmonics of the 10 cycles.
Figure 9: Torque response form simulation
From equation (3) the revolution per minute (rpm) of
the rotor in the self-running magnetic motor can be
calculated. The rpm is needed as it is part of the equation
(4) to find the mechanical power.
where
RCF = relative centrifugal force,
r = centrifugal radius in mm
From (3) the rpm can be plotted as shown in figure 10.
13
Figure 10: Revolution per minute (rpm) response
Once torque is obtained the mechanical power can be
calculated by using the equation shown in (4):
The main objective of this study and the most important is
the capability of the self-running magntic motor to induce
electricity. In normal cases the efficiency of the generator
are working around 90% to produce electrical power.
Hence, the electrical power can be derived from
mechanical power as shown in equation (5). The
comparison output of the mechanical power versus
electrical power are as shown in figure 11.
Electrical Power = 0.9 x Mechanical Power (5)
Figure 11: Mechanical power from self-running
magnetic motor
V. CONCLUSION
From the study, it can be concluded it is possible to
induce electricity from self-running magnetic motor.
However, this primarily finding will needs further
investigation before a prototype can be developed.
REFERENCES
[1] https://www.greenoptimistic.com/tesla-free-
energy/#.WgEzRVNx11s (7Nov2017)
[2] https://www.nuenergy.org/nikola-tesla-radiant-energy-system/
(7Nov2017)
[3] M. Casis, et al., "Free-Energy Generator," Pulsar, vol. 1, 2013
[4] https://www.duramag.com/techtalk/halbach-arrays/how-are-
halbach-arrays-designed/ (16Nov2017)
[5] J. Mallinson, “One-sided fluxes – A magnetic curiosity?” IEEE Transactions on Magnetics, vol. 9, no. 4, pp. 678–682, Dec. 1973
[6] Amel Ridha, “Design and Simulation of Free Energy Permanent
Magnet Motor (FEPMM)”, European Journal of Scientific
Research, vol. 138 No 3 March 2016, pp.123-132
[7] C.G.C. Neves, A.F.F. Filho, “Analysis a Magnetic Gear Integrated
Halbach PM Generator”, XVII International Symposium on Electromagnetic Fields in Mechatronics, Electrical and Electronic
Engineering (ISEF2015), Valencia, Spain, 10-12 September 2015.
14
A Review on Voltage Wave Reflection in
Transmission Lines
Awadh.Al-Kalbani Suhar College of Applied Sciences
Abstract—The paper starts with the traditional literature
theory on Telegrapher equations solution and wave propagation
and reflection along the transmission line. Few books were
referenced on this theory. In the discussion section, the paper lists
how the reflection theory is physically wrong. In the end of the
discussion, the paper argues, the wave in transmission line can be
decomposed in different ways and is best described as a signal
being modulated by the transmission line impedance.
Transmission line problems can be solved with physically correct
theory and basic circuit theory laws.
Index Terms— Reflection coefficient; Telegrapher equations;
Transmission lines; Voltage reflection.
I. INTRODUCTION
Books explain the wave propagation in TEM transmission
lines (TL), e.g. the coaxial cable, as a voltage (or current)
wave propagating and reflecting along the TL. This paper is to
lay the option of using physically correct and less confusing
theory in solving TL problems. Correct analysis of a
transmission line problem, can lead to a new solutions.
II. LITERATURE REVIEW
Figure 1: Lumped-element circuit model
A. TEM Transmission line model
A high frequency transmission line can be modeled by the
“lumped-element circuit” model as shown in Figure 1. The
Transmission line is orienting along the z-direction, Which is
subdivided into differential sections each of length Δz.
Where:
R': The combined resistance of both conductors per
unit length, in Ω/m.
L': The combined inductance of both conductors per
unit length, in H/m.
G’: The conductance of the insulation medium
between the two conductors per unit length, in S/m, and
C’: The capacitance of the two conductors per unit
length, in F/m.
B. Relationship between current and voltage in a
transmission line
Referring to Figure 1, and analyzing one section as shown in
Figure 2:
Figure 2: A section model
The solution of the relationship between the current and the
voltage as Δz→0 , is the Telegrapher equatios:
− 𝜕𝑣(𝑧, 𝑡)
𝜕𝑧= 𝑅′𝑖(𝑧, 𝑡) + 𝐿′
𝜕𝑖(𝑧, 𝑡)
𝜕𝑡
(1)
and :
−𝜕𝑖(𝑧, 𝑡)
𝜕𝑧= 𝐺′𝑣(𝑧, 𝑡) + 𝐶′
𝜕𝑣(𝑧, 𝑡)
𝜕𝑡
(2)
which is in phasor form:
15
−𝑑(𝑧)
𝑑𝑧= (𝑅′ + 𝑗𝜔𝐿′)𝐼(𝑧),
−𝑑𝐼(𝑧)
𝑑𝑧= (𝐺′ + 𝑗𝜔𝐶′)(𝑧).
(3a)
(3b)
the solution to these differential equations is:
(𝑧) = 𝑉0+𝑒−𝛾𝑧 + 𝑉0
−𝑒𝛾𝑧
𝐼(𝑧) = 𝐼0+𝑒−𝛾𝑧 + 𝐼0
−𝑒𝛾𝑧
(4a)
(4b)
where: Vo+
, Io+
, Vo- , Io
- , are constants but are called wave
amplitudes in the traditional TL theory, and:
𝛾 = √(𝑅′ + 𝑗𝜔𝐿′)(𝐺′ + 𝑗𝜔𝐶′) (5)
This is a constant and is called the complex propagation
constant.
γ can be split into real and imaginary parts:
γ=α+jβ (6)
C. Relationship between the current I (z) and the voltage
V (z)
Using equations 3a, and 4a, the following can be produced:
𝐼(𝑧) =𝛾
𝑅′ + 𝑗𝜔𝐿′[𝑉0+𝑒−𝛾𝑧 − 𝑉0
−𝑒𝛾𝑧] (7)
or rearranging:
𝐼(𝑧) =𝑉0+
𝑍0𝑒−𝛾𝑧 −
𝑉0−
𝑍0𝑒𝛾𝑧
(8)
where:
𝑍0 =𝑅′ + 𝑗𝜔𝐿′
𝛾= √
𝑅′ + 𝑗𝜔𝐿′
𝐺′ + 𝑗𝜔𝐶′ (𝛺)
(9)
D. Transmission line circuit
A common Transmission line configuration is shown in
figure 3.
Figure 3: Common transmission line circuit
Notice z=0 is chosen where the load is. Then d=-z, can be
used in the TL voltage and current equations.
A common assumption in TL circuits is α=0, then γ
becomes: γ=jβ
Thus the TL voltage and current equations would become:
(𝑧) = 𝑉0+𝑒−𝛽𝑧 + 𝑉0
−𝑒𝛽𝑧 ,
𝐼(𝑧) =𝑉0+
𝑍0𝑒−𝛽𝑧 −
𝑉0−
𝑍0𝑒𝛽𝑧
(10a)
(10b)
According to the traditional text:
“Vo+
, Io+
, Vo- , Io
- are unknown constants which can be
found in the context of the complete circuit.
𝑒−𝑗𝛽𝑧 is associated with a wave traveling in the positive z
direction, from the source (sending end) to the load (receiving
end). Accordingly, we will refer to it as the incident wave, with
Vo+
as its voltage amplitude. Similarly, the term containing
𝑒𝑗𝛽𝑧 represents a reflected wave with voltage amplitude Vo- ,
traveling along the negative z-direction, from the load to the
source.“ [1]
The same explanation from the next book:
“where Vo+
, Io+
, Vo- , and Io
- are wave amplitudes; the +
and — signs, respectively, denote
wave traveling along +z and –z directions, as is also indicated
by the arrows.”[2]
And this is the same explanation from another book:
“In considering the voltage function that will satisfy (13), it
is most expedient to simply state the solution, and then show
that it is correct. The solution of (13) is of the form:
V(z, t) = f1 (t – z/ ν) + f2 (t + z/ ν ) = V + + V –
where ν, the wave velocity, is a constant. The expressions (t
± z/ν) are the arguments of functions f1 and f2. The identities
of the functions themselves are not critical to the solution of
(13). Therefore, f1 and f2 can be any function. The arguments
of f1 and f2 indicate, respectively, travel of the functions in the
forward and backward z directions. We assign the symbols V
+ and V − to identify the forward and backward voltage wave
components. To understand the behavior, consider for
example the value of f1 (whatever this might be) at the zero
value of its argument, occurring when z = t = 0. Now, as time
increases to positive values (as it must), and if we are to keep
track of f1(0), then the value of z must also increase to keep
the argument (t − z/ν) equal to zero. The function f1 therefore
16
moves (or propagates) in the positive z direction. Using
similar reasoning, the function f2 will propagate in the
negative z direction, as z in the argument (t + z/ν) must
decrease to offset the increase in t. Therefore we associate the
argument (t − z/ν) with forward z propagation, and the
argument (t + z/ν) with backward z travel. This behavior
occurs irrespective of what f1 and f2 are. As is evident in the
argument forms, the propagation velocity is ν in both cases.”
[3]
The following things follow the above assumptions:
- The reflection coefficient at the load, which is used to
measure the energy transfer to the load:
Γ= Vo-/ Vo
+ (11)
best energy transfer to the load happens when Γ=0.
- The characteristic impedance, which is said to give
the relationship between the voltages and the
currents. 𝑉0+
𝐼0+ = 𝑍0 =
−𝑉0−
𝐼0− (Ω) (12)
E. Another term for the relationship between the current
I (z) and the voltage V (z)
𝑍(𝑧) =(𝑧)
𝐼(𝑧)=𝑉0+𝑒−𝛾𝑧 + 𝑉0
−𝑒𝛾𝑧
𝐼0+𝑒−𝛾𝑧 + 𝐼0
−𝑒𝛾𝑧
= 𝑍0(
𝑉𝐿𝐼𝐿+ 𝑍0𝑡𝑎𝑛ℎ[𝛾(𝑙 − 𝑧)]
𝑍0 +𝑉𝐿𝐼𝐿𝑡𝑎𝑛ℎ[𝛾(𝑙 − 𝑧)]
)
(13)
III. OBJECTIVES
- To correct the traditional theory of voltage reflection in
transmission lines.
- Provide a correct explanation to the voltage wave on the
transmission line which may be followed by a corrected
approach in solving TL problems, e.g. matching.
IV. DISCUSSION
A. Reflecting wave don’t occur for the following reasons:
i. Equations 10a and 10b happen to be a
mathematical solution to differential equations.
Both Vo+ and Vo- are obtained from V(0) and
V’(0) (when getting the particular solution to a
differential equation) and not from a V(z) at the
reflection point. Therefore, V(z) can be
decomposed into to sinusoidal waves one a
function of z and the other a function of –z. It
can also be decomposed into different ways not a
function of (-z). Therefore, there is no reason to
extend the mathematical solution into 𝑉0− being a
result of some sort of a physical reflection.
ii. It breaks circuit theory law (i.e. what’s normally
called Ohms law for phasors): 𝑉(𝑧)
𝐼(𝑧) = Z(z) , should always be the case to stasify circuit
theory laws.
However:
𝑉0+
𝐼0+ = 𝑍0 =
−𝑉0−
𝐼0−
does not satisfy circuit theory laws. 𝑍0is generally not the
impedance of the TL.
iii. The Electric Field and the Magnetic Field do not
travel along the TL, that’s why the TL is referred
to as TEM line. e.g. in the coaxial cable case, the
Electric Field travels from the center conductor
to the outer conductor across the cable. Thus the
current following in a TL should not be confused
with the Electromagnetic (EM) wave
propagation. i.e. The EM waves do not travel
from the source to the load. The current travels
from the source to the load.
iv. The theory of the reflecting wave is not
repetitive, units have to be made up and require a
lot of fix’s for different cases. e.g.:
o The reflection coefficient at the load is not
the same throughout the line. Γ(z) depends
on the distance on the line! How this can be
comprehended physically?
o The relationship between the incident
voltage amplitude Vo+
and incident current
amplitude Io+ is 𝑍0 as shown in equation
(12), but the relationship between the
reflected voltage amplitude and the reflected
current amplitude is Io- is –𝑍0 (i.e. with a
negative sign).
o 𝑍0 in equation (9) is assigned to have a unit
of ohms, but it doesn’t really calculate to a
unit of ohms. R’ unit is ohms per meter.
B. An alternative solution to the Telegrapher equations
By substituting equation 3a and 3b into each other:
𝑑2(𝑧)
𝑑𝑧2− 𝛾2(𝑧) = 0
𝑑2𝐼(𝑧)
𝑑𝑧2− 𝛾2𝐼(𝑧) = 0
(14a)
(14b)
For a lossless TL, γ=jβ, therefore equation (14a) becomes:
𝑑2(𝑧)
𝑑𝑧2+ 𝛽2(𝑧) = 0
(15)
Applying Laplace transform to solve the equation:
𝑠2𝑉(𝑠) − 𝑠(𝑧 = 0) − ′(𝑧 = 0) + 𝛽2𝑉(𝑠) = 0
Rearranging:
17
𝑉(𝑠)[𝑠2 + 𝛽2] = 𝑠(𝑧 = 0) + 𝑉′(𝑧 = 0) Therefore:
𝑉(𝑠) = 𝑠(𝑧 = 0) + 𝑉′(𝑧 = 0)
[𝑠2 + 𝛽2]
= 𝑠(𝑧 = 0)
[𝑠2 + 𝛽2]+ 𝑉′(𝑧 = 0)
[𝑠2 + 𝛽2]
Now taking inverse Laplace transform:
(𝑧) = (𝑧 = 0) cos𝛽𝑧 +𝑉′(𝑧 = 0)
𝛽sin 𝛽𝑧
Where (𝑧 = 0) 𝑎𝑛𝑑 𝑉′(𝑧 = 0) are generally
complex values. (𝑧) in equation (16) is readily
decomposed into two waves one is (𝑧 = 0) cos𝛽𝑧
And 𝑉′(𝑧=0)
𝛽sin 𝛽𝑧. Now writing it in time domain:
𝑣(𝑧, 𝑡) = |(0)| cos(𝛽𝑧) 𝑐𝑜𝑠 (𝜔𝑡 + ∠(0)) +
|𝑉′(0)|
𝛽𝑠𝑖𝑛(𝛽𝑧)𝑐𝑜𝑠 (𝜔𝑡 + ∠𝑉′(0))
Equation(17), shows that the transmitted wave
(analogous to carrier) would be modulated by the TL
(or (𝑧)) , instead of being reflected.
To find |(𝑧)| which is detectable by SWR meter, lets
first rewrite (𝑧) 𝑎𝑠:
(𝑧) = (𝑎 + 𝑗𝑏) cos 𝛽𝑧 +(𝑐 + 𝑗𝑑)
𝛽sin 𝛽𝑧
Where, a+jb=(𝑧 = 0) and c+jd= 𝑉′(𝑧 = 0):
Then:
(𝑧) = 𝑎 cos𝛽𝑧 +𝑐
𝛽sin 𝛽𝑧
+ 𝑗 [(𝑏 𝑐𝑜𝑠 𝛽𝑧 +𝑑
𝛽𝑠𝑖𝑛 𝛽𝑧]
Then using Pythagoras law for magnitude and basic
trigonometry for the angel:
(𝑧) =
√[𝑎 cos 𝛽𝑧 +𝑐
𝛽sin𝛽𝑧]
2
+ [(𝑏 𝑐𝑜𝑠 𝛽𝑧 +𝑑
𝛽𝑠𝑖𝑛 𝛽𝑧]
2
tan−1𝑏 𝑐𝑜𝑠 𝛽𝑧 +
𝑑𝛽𝑠𝑖𝑛 𝛽𝑧
𝑎 𝑐𝑜𝑠 𝛽𝑧 +𝑐𝛽𝑠𝑖𝑛𝛽𝑧
Then, |(𝑧)| =
√[𝑎 cos 𝛽𝑧 +𝑐
𝛽sin𝛽𝑧]
2
+ [(𝑏 𝑐𝑜𝑠 𝛽𝑧 +𝑑
𝛽𝑠𝑖𝑛 𝛽𝑧]
2
(16)
(17)
(18)
=
√ [𝑎2 + 𝑏2
2−𝑐2 + 𝑑2
2𝛽2] cos(2𝛽𝑧)
+ [𝑏𝑑 + 𝑎𝑐
𝛽] sin(2𝛽𝑧)
+ [𝑎2 + 𝑏2
2+𝑐2 + 𝑑2
2𝛽2]
If the load is matched |(𝑧)| should become a
straight line against z (as known in TL). So matching
conditions are:
[𝑎2 + 𝑏2
2−𝑐2 + 𝑑2
2𝛽2] = 0 𝑎𝑛𝑑 [
𝑏𝑑 + 𝑎𝑐
𝛽] = 0 𝑜𝑟 𝑏𝑑
= −𝑎𝑐
But 𝑎2 + 𝑏2 = |𝑉 (0)| 2and 𝑐2 + 𝑑2 = |𝑉′(0)|2
Thus the matching conditions are:
|(0)| 2 =|𝑉′(0)|2
𝛽2 𝑜𝑟 |(0)| =
|𝑉′(0)|
𝛽𝑎𝑛𝑑
𝑏
𝑎= −
𝑐
𝑑 𝑦𝑖𝑒𝑙𝑑𝑠→ ∡(0) = ∡[′(0)]
Combining the two conditions:
(0) =′(0)
𝛽
Which happens if the load impedance matches with
Thevenin impedance of the circuit. This statement is
alternative to the traditional theory of reflection
coefficient Γ=0.
V. CONCLUSION
- The voltage reflection theory is not correct though it
works (with few fixes) to solve TL problems.
- The voltage reflection theory is not essential to solve TL
problems. They can be solved using correct
understanding of standing wave and basic circuit theory
analysis.
- A complete corrected theory and methods can be
developed to update the traditional methods for solving
TL problems. e.g. more work can be done to write the
relations of ZL, Zo, (0)𝑎𝑛𝑑 𝑉 ′(0).
VI. REFERENCES
[1] Fawwaz T. Ulaby, Eric Michielssen. Fundamentals of Applied
Electromagnetics. Sixth. London : Pearson, 2014. pp. 54-78.
[2] N.O.Sadiku, Mattew. Elements of Electromagnetics. s.l. : Oxford, 2011. p. 515.
[3] William H. Hayt, J. John A.Buck. Engineering Electromagnetics. s.l. :
McGraw Hill, 2006. p. 307.
18
Development of Dual Axis Solar Cells Tracking
Prototype Based on PVC Foam Material
H. Sharabaty, O. Khazraji Department of Electrical & Electronics Engineering, University of Turkish Aeronautical Association, Ankara, Turkey
Abstract—with a view to evolving additional solar energy and to
raise the efficiency of the solar photovoltaic panels, tracking
systems are used to follow the sun position and to let the solar
photovoltaic panels head for the sun as long as possible. This paper
proposes to use PVC foam material instead of the other materials
such as Iron and Plastic in order to develop a dual axis solar panel
tracking prototype based on AVR microcontrollers. Compared to
other materials, PVC foam withstands more temperatures and can
reduce the design weight and power consumption. In addition, this
work proposes to discretize the tracking operation of sun position
by switching DC relay, and to study the effect of the consumed
energy, by the control circuit and motors, on the total generated
energy of the system, and then to determine the optimal DC relay
timing period. Finally, this work discusses the advantages of
connecting solar cells in series regarding the solar panel output
voltage. The experimental work shows that compared to the static
solar system, the proposed dual axis solar tracking prototype is
more efficient for absorbing higher sunlight by increasing the
average power by 34%. Moreover, the measurements proved that
the optimal interval for discretize tracking system is 20 minutes
that gives the maximum achieved efficiency of the prototype, i.e.
34%. Compared to the continuous tracking system, the optimal
tracking interval saves 89.45% of the consumed energy by the
control circuit.
Index Terms— AVR microcontrollers; Discrete sun tracking;
Dual axis solar tracker; Solar cells; PVC foam material.
I. INTRODUCTION
Solar energy is very effective and efficient to produce
electricity especially in countries where the warm climate is
mostly dominant [1]. The problem here is that the solar
photovoltaic (PV) has low efficiency in producing maximum
output power which is derived from sunlight [2]. To fix this
problem up, several works have been achieved to increase the
efficiency of the output power by using the solar tracking
system that maintains the solar PV panels perpendicular to sun
radiation [2]. This can increase the efficiency of energy
production up to 35% based on annual estimation [3]. The first
proposed solar tracking systems were single axis tracker [4, 5,
6]. Later on, the most efficient algorithm to control the dual-
axis solar tracker which could rotate in direction of azimuth and
elevation were simulated and implemented [7, 8, 9, 10]. The
previous studies show that two types of solar panel tracking
prototypes were proposed:
The first one contains big solar panels and can be used for
energy-generating purposes because the total average of the
generated energy is higher than the total energy of the
consumption [11, 12, 13].
The second type was dedicated for research purposes in labs
and universities and was contains small solar panels. The main
problem here is that the total energy of the consumption of these
small prototypes is higher than the total average of the
generated energy [6, 8, 9].
II. METHODOLOGY AND PROPOSED DESIGN
This part focuses on the design and implementation of the
proposed dual axis solar panel tracker prototype. The
mechanical parts of the proposed system are drawn by
AutoCAD 2017 program because it is considered as the main
part of utilizing the PVC foam in the Computer Numerical
Control machine (CNC). By using the AutoCAD to draw the
required pieces, the commands and the drawings sent to the
PVC foam cutting machine (CNC) in order to cut the required
pieces. Later on, we put Length = 1 meter & Thickness = 1cm
from the PVC foam material under the CNC machine and turn
on the cutting machine. Directly, the CNC started to cut
according to the design, drawings, and scales. The aim of the
proposed dual axis solar panel tracking system is to keep the
solar panel facing toward the sun all the day. Therefore, we
have to track the position of the high intensity of the sunlight
by using the LDRs. The proposed design depends on the
obtained value of the voltage difference between the four LDRs
which are located in a shape of a square on the upper section of
the design. Automatically, the panel will rotate according to the
obtained value. As shown below in figure 1, the proposed dual
axis solar panel tracking system consists of the sensing unit, the
microcontroller unit, and the two servo motors (rotation unit).
Figure 1: Block diagram of the proposed prototype
The components of the proposed solar panel tracking system
are very sensitive. Therefore, it requires precision and
concentration during the work. These components are
19
electrical, electronic and mechanical. The proposed solar panel
tracking system consists of two main circuits. The purpose of
the first circuit is to sense the solar radiation by using LDRs, to
control the Arduino Uno which is the brain of the design and to
control the two servo motors (vertical and horizontal). The
purpose of the second circuit which is the solar panel circuit is
to convert the solar energy into electrical energy and to supply
the electrical energy into the load.
A. Arduino Uno board is the microcontroller and the brain of
our design. It receives analog signals from the LDRs by the
analog pins and sends pulse width modulation signals (PWM)
by the PWM digital pins to the two servo motors (horizontal
and vertical). Therefore, regarding the voltage of this
microcontroller unit, the Input voltage is 9V.
B. Servo motor (MG 946R) is the mechanical component of
the proposed design. It has tough metal gears. We preferred this
type of servo motors because it has high quality and high torque
during the work of the solar panel tracking system. It has three
main wires, the orange wire is for the pulse width modulation
signal (PWM), the red wire is for the operating voltage (V+)
and the last wire is for the ground. There are two servo motors
which are the horizontal and the vertical servo motors. The
horizontal servo motor controls the horizontal angle of the
design (from east to west) while the vertical servo motor
controls the vertical angle of the design (from south to north).
Eventually, the operating voltage for the two servo motors is
7.2V.
C. The light dependent resistor (LDR) is the sensing part of
the proposed design. This part can work as a light sensor to
detect the absence and the presence of light. We chose them
because they can cost a low price, as well as it has high
sensitivity and simple structure. We put four LDRs in the upper
section of the design in a shape of a square. Directly as the
(LDRs) sensing the light, they will give analog signals to the
microcontroller. Moreover, the input voltage to the LDRs is 5V.
The fixed resistors which are connected in series with the LDRs
to form a potential divider circuit. The center point of the
potential divider is fed to the analog pins of the Arduino board.
In practical, we selected four fixed resistances in the proposed
design and the value of each resistance is (360Ω).
D. Voltage regulation (LM2596) is a DC to DC step down
regulation. We chose it because it is adjustable and ideal for the
proposed design and we needed to step down the input voltage
from 9V (adaptor) to 7.2V (servo motor voltage). We regulated
the voltage from the adjustable part by using the screw-driver
and checked the voltage by using the AVO meter.
E. The thermometer is used to measure the temperature of the
solar panel. The unit of this part is Celsius degree. It is very
accurate to the exact temperature. The input voltage for this part
is 5V.
F. We chose a resistance (10W 300ΩJ) as a load which is
connected with the solar cells (series or parallel case). The load
voltage and the current depend on the solar energy absorbed by
the solar panel. Eventually, we measured the output voltage and
the current on both ends of this load by using the AVO meter.
G. 10 solar cells (6V 0.1W) in two cases (parallel and series
connection) cost low prices and give high efficiency in
absorbing the solar energy and converting it into electrical
energy. Even though the output power of the solar panel is quite
small, but the solar panel is enough to show that the sunlight
energy can be grabbed as much as possible because the solar
panel is moving in response to the direction of sunlight that
sensed by LDRs.
III. THE WORKING PRINCIPLE
The working principle of the proposed prototype is when the
sunlight falls on the LDRs, the microcontroller will sense the
variation of light by the analog pins. We have 3 cases:
A. Case 1
When the sun is located on the right or left or up or down the
four LDRs, the voltage value will increase in the analog pins
which are connected on the LDRs toward the sun and will
decrease on the LDRs opposite to the sun. For example, if the
sun is located on the right side of the four LDRs, the voltage
value will increase in the right analog pins and will decrease in
the left analog pins. Then, the microcontroller will calculate the
average of the increasing voltage between the right up LDR and
the right down LDR. Also, it will calculate the average of the
decreasing voltage between the left up LDR and left down
LDR. By applying the subtraction operation on the two
resulting averages, the resulting value from the operation will
take these results as commands to move the horizontal motor to
the right direction and to equalize the voltages of the analog
pins by equalizing the intensity of the sunlight on the LDRs.
Moreover, for the other directions, we can apply the same
operation in order to show the aimed results.
B. Case 2
When the sun is located on the right up or left up or right down
or left down the four LDRs i.e. the effect of the sun will be in
two directions (will not be in one direction). For example, if the
sun is on the right up of the four LDRs, the LDR which is
located on the right upside will be towards the sun and the
voltage of this analog pin will increase. Contrastingly,
regarding the sensor which is located on the opposite side to the
sun (the left down sensor), the voltage of the connected analog
pin will be lower than the other pins. Thus, when the
microcontroller does the calculation, the right analog pins
voltages will be increased and the upper analog pins voltages
will also be increased. Moreover, the vertical servo motor will
move up and the horizontal servo motor will move to the right
till the voltages of the analog pins will be equalized. This
process can be applied to the other directions to show the aimed
results.
20
C. Case 3
In the case of darkness or the same amount of the falling solar
rays on the four LDRs, the voltages of the four analog pins will
be equal and the microcontroller will command the 2 servo
motors to stay in the same position (no motion).
We programmed the Arduino Uno board by using the Arduino
C language to write the codes and command the Arduino C
program (Personal Computer) to send these codes to the
microcontroller board by using external cable. The Flowchart
of the tracking operation is shown in Figure 2.
Figure 2: Flowchart of the tracking operation
The proposed dual axis solar panel tracking system, which has
been practically achieved, is shown in Figure (3).
Figure 3: The proposed prototype
IV. RESULTS AND DISCUSSIONS
In this part, we summarize the acquired results during our work.
We start by comparing our prototype with another prototype
developed by the department of electrical and electronic
engineering at Baghdad University. This prototype was built
for research purposes in Renewable energy laboratory and was
made of iron, plastic and cements materials. Then, we will
compare between the energy resulted from the static solar cells
system and from the dual axis solar tracker by using ten solar
cells with the same output power. We will discuss both cases:
series and parallel connection. After that, we will show the
results of the tracking and static systems of the parallel case and
calculate the average power. Then, we will display the energy
values of the tracking and static prototypes of the series case.
The tracking system will be track the sun by switching the DC
relay for an interval of time (5 mins, 10 mins, 15 mins, 20 mins,
30 mins, 40 mins, 50 mins, 55 mins, 1 hr, 2 hrs, 3 hrs, 4 hrs and
6 hrs) for a duration of 12 hrs while the static prototype will
remain fixed toward the sun at 90 degrees angle all the day. We
will determine the energy consumption and generation of our
prototype for all cases and through them, we will calculate the
efficiency. The measurements have been done in Baghdad
during the period from 10 to 31 August 2016, and the results
were registered according to specific intervals of time from 6
am to 6 pm because sunrise in that period was between 5:21 and
5:35 am, whereas the sunset was between 18:53 and 18:28 pm.
A. Weight of Our Proposed Design
We compared the weight of our prototype with a prototype built
at the University of Baghdad in the laboratory of electrical and
electronic engineering for research experiments of student's
research with the same specifications but made of iron, plastic
and cement as shown in the figure 4 and compare the weight
values of the prototypes illustrated in table 1.
Table 1
Comparison with reference prototype
THK Prototype UOB Prototype
Weight of The Prototype
2.5 Kg
4 kg
Material
PVC Iron+plastic+Cement
Figure 4: The reference prototype and our proposed prototype
Therefore, the proposed design has less weight than the other
design.
B. Parallel Solar Cells Connection
We will connect ten solar cells in parallel to the load to calculate
the average power value as shown in figure 5.
This figure demonstrates the data of power which are collected
by static and rotating solar panels for the First day. The static
solar panel data displays that the maximum power is 0.2045 W.
21
For the meantime, the rotating solar panel data displays that the
highest power is 0.1939 W. From figure 4, the power values
produced by the rotating panel are approximately equal during
the time of testing. But, for the static panel, it can be seen that
the solar intensity increases from 06:00 to 07:00 until 09:00 to
10:00. Then, the solar intensity starts to decrease smoothly until
at 17:00 to 18:00.
Figure 5: Interpretation data of power of solar panel 0.12 W
Figure 4 shows an increase in the total average power which is
collected from the tracker solar panel in comparison to the total
average power which is collected from the static solar panel.
C. Series Solar Cells Connection
We will connect ten solar cells in series case to the load to
calculate the average energy value by switching the DC relay
for multiple intervals of time (5 mins, 10 mins, 20 mins, 30
mins, 40 mins, 50 mins, 55 mins, 1 hr, 2 hrs, 3 hrs, 4 hrs, and 6
hrs) for a duration of 12 hours.
The first case is switching the DC relay by using an interval of
5 minutes for 12 hours
Figure 6: Rotating vs. static system for switching interval of 5mins
The results illustrated in figure 6 shows the obtained energy
from the static and the tracker panels. By using the tracking
system, the maximum energy was 237.4682 J could be reached
from 10:55 to 11:05. Contrastively, without using the solar
tracking system (static system), the maximum energy reached
from 12:00 to 12:05 is 178.47 J. Also, the obtained energy of
the tracker panel at any other time period is more than the
obtained energy of the static panel at the same period. Figure 6
also shows an increase in the total average energy which is
collected from the tracker solar panel in comparison with the
total average energy collected from the static solar panel.
Calculating the area under the envelope of the generated energy
curve for the solar tracking system gives us the total average
energy (Et) = 25871.50 J/Day. As a result, the total average
energy obtained from the solar tracking system that follows the
sun by an interval of 5 minutes for 12 hours is equal to 25871.50
– 10760.94 = 15110.56 J/day. On the other hand, the total
average of the generated energy from the solar static system
(Es) is equal to 11322.85 J/Day. So, we can define the
efficiency of our prototype by:
The efficiency = ((Et-Es)/Es) * 100% ………… (1)
Where
Et: The total average generated energy by the tracking system.
Es: The total average generated energy by the static system.
Using the equation 1, the efficiency of our system by using
switching interval of 5 minutes is 33.45%. These calculations
can be applied to the other cases to show the aimed results.
D. Energy Consumption of the Prototype
Here, we will consider that the continuous sun following system
will switch the DC relay using an interval of 1 second. The
Energy consumption of the control circuit is 10458.72 J and the
energy needed to move the motors for 1 second is 89424 J as
well as the motor needs energy to return to the start after the
demise of the sun by 4.14 J. Therefore, the total energy
consumption in the case of switching DC relay using an interval
of 1 second for 12 hours is 99886.86 J and Figure 7 explains the
total energy consumption for the other different switching
intervals.
00.05
0.10.15
0.20.25
06
:00
to
07
:00
07
:00
to
08
:00
08
:00
to
09
:00
09
:00
to
10
:00
10
:00
to
11
:00
11
:00
to
12
:00
12
:00
to
13
:00
13
:00
to
14
:00
14
:00
to
15
:00
15
:00
to
16
:00
16
:00
to
17
:00
17
:00
to
18
:00
Pow
er in
Watt
Time
Rotating Static
050
100150200250
06
:00
to
…
06
:55
to
…
07
:50
to
…
08
:45
to
…
09
:40
to
…
10
:35
to
…
11
:30
to
…
12
:25
to
…
13
:20
to
…
14
:15
to
…
15
:10
to
…
16
:05
to
…
17
:00
to
…
17
:55
to
…
En
ergy i
n J
Switching Interval Time
Rotating Static
10300
10350
10400
10450
10500
10550
10600
10650
10700
10750
10800
5 m
ins
10
min
s
20
min
s
30
min
s
40
min
s
50
min
s
55
min
s
1 h
r
2 h
rs
3 h
rs
5 h
rs
6 h
rsE
ner
gy C
on
sum
pti
on
in
J
Switching Interval Time
22
Figure 7: Energy consumption Curve
The highest total energy consumption appears in the case of
switching the DC relay by using an interval of 5 minutes for 12
hours while the less total energy consumption appears in the
case of switching the DC relay by using an interval of 6 hours
as shown in figure 7.
E. The Generated Energy Gain
We will consider that the difference between the total average
of the generated energy and the total consumed energy as
produced energy gain.
Figure 8: Generated energy gain curve
The highest generated energy gain appears in the case of
switching the DC relay by using an interval of 20 minutes for
12 hours while the less generated energy gain appears in the
case of switching the DC relay by using an interval of 6 hours
as shown in figure 8.
F. Efficiency of The Proposed Prototype
It is the ratio of the difference between the total average of the
generated energy of the tracking prototype and the total average
of the generated energy of the static prototype to the total
average generated energy of the static prototype.
The highest efficiency appears in the case of switching the DC
relay by using an interval of 20 minutes for 12 hours by 34% as
shown in figure 9.
Figure 9: Efficiency curve of the proposed prototype
V. CONCLUSIONS AND RECOMMENDATIONS
In this work, we designed and implemented a dual axis solar
tracker prototype to track the movement of the sun as the sun
rises from the east and sets into the west during the day. The
prototype has been designed by using AutoCAD 2017. The
mechanical parts have been cut by using a Computer Numerical
Control (CNC). The control circuit was achieved by using the
AVR microcontroller and it has been practically tested. We also
proposed to use PVC foam material in order to reduce the
weight of the prototype and decrease the energy consumption.
Likewise, PVC foam material can withstand the high
temperatures of Iraq weather that can arrive at 56. For
decreasing the energy consumption and increasing the
efficiency of the prototype, we propose the discretizing for the
tracking operation of sun position by using a switching DC
relay. The optimal switching interval was 20 minutes that
increased the efficiency of our prototype to 34%. This optimal
tracking period saved 89.45% of the requested energy for the
control circuit in comparison with the continuous tracking
system. Our proposed dual axis solar tracking prototype is more
advantageous for capturing the maximum sunlight with
increasing the average power by 34% in comparison with the
static solar system which using the optimal switching interval. According to the results and the experiments in this work, many
suggestions can be proposed to develop the tracking design for
the future such as studying the optimal switching interval for
larger solar panels. Also, replacing the control circuit of our
prototype system with another control circuit can consume less
energy such as STM32 Nucleo-64 development board with
STM32F401RE MCU.
REFERENCES
[1] A. Zakariah, J. Jamian, M. A. M. Yunus, “Dual-axis solar tracking system
based on fuzzy logic control and Light Dependent Resistors as feedback
path elements,” in 2015 IEEE Student Conference on Research and
Development (SCOReD), 2012.
[2] C. Alexandru, “The design and optimization of a photovoltaic tracking
mechanism,” in 2009 International Conference on Power Engineering, Energy and Electrical Drives, 2009, pp. 436-441.
[3] M. A. Usta, Ö. Akyazi, İ H. Altaş, “Design and Performance of Solar
Tracking System with Fuzzy Logic Controller,” in 6th International Advanced Technologies Symposium (IATS’11), Elazığ, Turkey, 2011, pp.
381-385.
[4] T. Tudorache, L. Kreindler, “Design of a Solar Tracker System for PV Power Plants,” Acta Polytechnica Hungarica, vol.7, 2010, pp. 23-39.
0
5000
10000
15000
20000
5 m
ins
10
min
s
20
min
s
30
min
s
40
min
s
50
min
s
55
min
s
1 h
r
2 h
rs
3 h
rs
4 h
rs
6 h
rsGen
erate
d E
ner
gy
Gain
in
J
Switching Interval Time
32.50%
33.00%
33.50%
34.00%
34.50%
5mins
10mins
20mins
30mins
40mins
50mins
55mins
1 hr
Eff
icie
ncy
%
Switching Interval Time
23
[5] A. Dolara, F. Grimaccia, S. Leva, M. Mussetta, R. Faranda, M. Gualdoni,
“Performance Analysis of a Single-Axis Tracking PV System,” IEEE
Journal of Photovoltaics, 2(4), 2012, pp. 524-531.
[6] A. R. Waheed, “Implementation of solar energy tracking system using microcontroller (Unpublished master's thesis),” Electrical Engineering,
University of Technology-Iraq, 2013.
[7] S. Ozcelik, H. Parkash, and R. Challoo, “Two-axis solar tracker analysis and control for maximum power generation,” Procedia Computer
Science, vol. 6, pp. 457–462, 2011.
[8] A. A. Bin Azman, “A solar tracking system with multiple input parameters for efficiency optimization (master's thesis),” Faculty of
Electrical Engineering, Universiti Teknologi Malaysia, 2014.
[9] M. A. Bin Shukor, “Design of low power automatic sun tracking system using arduino uno (master's thesis),” Faculty of Electrical Engineering,
Universiti Teknologi Malaysia, 2015.
[10] K. Vijayalakshmi, B. Narendra, K. S. Anjaneyulu, “Designing a Dual Axis Solar Tracking System for Maximum Power,” Journal of Electrical
& Electronic Systems, 5(3), 2016, pp. 1.
[11] D. R, B. V, R. R, P. A, D. S, M. P, “Comparison of Efficiencies of Solar Tracker systems with static panel Single Axis Tracking System and Dual-
Axis Tracking System with Fixed Mount,” International Journal of
Engineering and Technology (IJET), 5(2), 2013, pp. 1925-1933. [12] H. S.Akbar, A. I. Siddiq, M. W Aziz, “Microcontroller Based Dual Axis
Sun Tracking System for Maximum Solar Energy Generation,” American
Journal of Energy Research, 5(1), 2017, pp. 23-27. [13] B. Krishna, “Tracking of Sun for Solar Panels and Real Time Monitoring
Using LabVIEW,” Journal of Automation and Control Engineering, 1(4), 2013, pp. 312-315.
24
Optimum Coordination of using Overcurrent Relay
using Two Phase Simplex and Ant Colony
Optimization Algorithm
Cheyaden Savio Aswin1, Dr.O.V.Gnana Swathika1,* VIT University Chennai, India
Abstract— Power Systems are prone to damage due to
overcurrent which is a result of faults like ground faults, line
faults, short circuit etc. To minimize the damage caused by these
faults suitable protection systems must be in place. The
protection systems consist of a primary system and a backup
system with proper coordination between the two systems (i.e.
their tripping time) in order to ensure proper clearance of faults
in minimum time. In this paper, the optimization of overcurrent
relays which are used primarily as backup systems against these
faults are analyzed. The application of Ant Colony Optimization
and Two Phase Simplex Algorithm in radial system is done to
obtain the time multiplier settings of the relays. This enables us to
achieve proper coordination between the overcurrent relays in
the network.
Index Terms— Ant Colony Optimization, Protection, Radial
Distribution, Swarm Intelligence, Two Phase Simplex Method.
I. INTRODUCTION
Electric Power is transmitted to the consumers from
generation centers through distribution systems. The electric power needs to be transmitted at low voltage levels to minimize loss and is stepped down at substation. Using primary feeders this stepped down electric power is fed to the distribution transformers. The type of distribution used depends on location and economics, but it is easier to coordinate current based devices if they are in a radial network [1-4].
A radial network consists of one power source and group of customers in series. All the customers are affected if there is a power failure. In order to minimize the damage to the system and interruption of power supply the importance of reliable protective systems is paramount and this is done keeping in mind that the occurrence of abnormalities in power systems is unavoidable [5-10]. Distribution systems in general, have two lines of defense a primary protection system and a backup system. The primary system acts as the first line of defense against faults the backup system comes into play in case of failure of the primary system [11-14]. The backup system would operate only after a certain period of time known as Coordination Time Interval (CTI) in order to give a chance to the primary system to operate [15-17]. The primary system consists of overcurrent relays. With the help of current and voltage transformers, the relays detect faults. The settings
of the relay must be done in a way that the relay located closest to the fault should have the minimum time of operation. Nevertheless, complex situations may arise which can lead to the faulty operation of relays. Hence, optimum coordination between relays is necessary [18].
This paper implements the use of Swarm Intelligence
Algorithm namely Ant Colony Optimization and Two Phase
Simplex Optimization to find the optimum Time Multiplier
Settings (TMS) of the relays in order to ensure minimum time
of operation of relays.
II. OVERCURRENT RELAY COORDINATION OF A
TWO-BUS RADIAL SYSTEM
There are two types of overcurrent relays: directional and non-directional relays. As Directional overcurrent (DOC) relays, do not require coordination with the relays behind them, they are preferred over the non-directional relays.
Figure1 : A radial two bus system
A radial feeder with two sections and feeders is shown in Fig 1. For a fault at F, relay R2 will be the first to operate. R2 operates after 0.1 sec time after the inception of the fault in order to protect the relay from transient current surges in the relay. Relay R1 should operate after a fixed time interval, CTI, which is equal to the sum of operation time of circuit breaker at bus 2, overshoot time of relay R2 and 0.1 sec. Similarly, these conditions can be expanded to larger networks. Using these constraints, the system is formulated as a Linear Programming Problem (LPP) and solution is obtained using the Two Phase Simplex and Ant Colony Optimization Algorithm. In this manner, we obtain the TMS and time of operation of relays R1 and R2.
25
A. Problem Formulation
DOC relays require two main parameters to operate
namely relay current settings and the time TMS [8]. Relay
settings depend on the maximum load current in the feeder.
TMS is obtained by minimizing the objective function [9-15]:
n
Min z = Σ topi (1)
i=1
where,
topi operating time of the primary relay i, for a fault at i
under the following constraints [2]:
B. Bounds on Operating Time –
topimin topi topimax (2)
where,
topimin the minimum time required for operation of
the relay at i for fault at ‘i’.
topimax time required for operation of the relay at i
for a fault at ‘i’.
C. Coordination Time Criteria –
Coordination time is the minimum time required
between operation of two relays [2].
tbopi -- topi ≥ Δt (3)
where,
tbopi - the operating time of the backup relay i, for a
fault at ‘i’.
Δt - the coordination time interval (CTI) [2].
D. Relay Characteristics –
Normal inverse definite minimum time (IDMT)
characteristics are assumed for all relays [2,5].
𝛼= 𝜆 (4)
(𝑃𝑆𝑀)𝛾−1
where,
λ is 0.14and is 0.02.
Plug multiplier setting (PSM) is given by
PSM = If (5)
CT ratio x Relay Setting
where,
If is the fault current (in A).
topi = λ ∗ (TMS) ∗ ((PSM)γ– 1) −1 (6)
i.e. topi = α(TMS) (7)
Substituting (7) in (1) gives the objective function as:
n
Min z = Σ αi(TMS)i (8)
i=1
The value of TMS is hence determined.
III. TWO PHASE SIMPLEX ALGORITHM
The two phase method is used to solve a linear
programming problem. It is used to retain optimality while
bringing the primal set of equation back to feasibility. It is
useful for re-optimizing a problem after a constraint is added
into a problem or some parameters of the same are changed so
that the previous optimal basis remains no longer feasible [4].
The algorithm is [4, 5]:
A. Algorithm
1. Start.
2. Try and convert the linear programming problem in
maximization form.
3. Check if all constraints are in ≥ form, if not then
convert them into the same.
4. Introduce slack variables to remove inequalities and
transform them into equalities.
5. Create a table by considering artificial coefficients as
basis variables.
6. Initialize the Cq values of non basis variables by
comparing it from the given equation.
7. Now fill up the table by entering all the values of
basis as well as non basis variables and also the RHS
column form the given constraint equations.
8. The Zq values of all the non basis variables are
calculated by the summation of product of cost and
the corresponding non basis values there after
calculate the values of Zq-Cq.
9. Now check whether all the values of Zq-Cq are
positive or not, if so then stop the process.
10. The column having most negative value of Zk-Cq is
taken as key column and the corresponding column
26
variable will be treated as the one that enters the
basis.
11. The values in the RHS column are divided by the
values in the corresponding key column for each row.
The row having minimum such values will be taken
as key row. The values obtained if found to be
negative will not be considered as key row.
12. The row having minimum value obtained from the
previous step, the variable corresponding to that row
leaves the basis.
13. The element corresponding to key row and key
column is taken as pivot element.
14. Make pivot element as one and make the
corresponding elements as zero by using row
transformation method in order to obtain a modified
table.
15. Develop the next improved solution by repeating the
process till all Zq-Cq becomes non-negative.
16. Repeat the same process for the second phase
iterations with the only difference in costs that are
taken as original coefficients of objective function,
until all the values of Zq-Cq becomes non negative.
17. The right hand side (RHS) column values obtained
in the final step gives the optimized solution.
18. end
IV. ANT COLONY OPTIMIZATION ALGORITHM
The Ant Colony Optimization(ACO) Algorithm uses the behavior of forging ants to determine the optimum solution of a problem. While foraging for food ants tend to distribute over an area to speed up the process. To indicate a path has been explored, each ant secrets pheromones while travelling. Thus the pheromone concentration for the most travelled path increases when the ants find the food source and the paths begin to overlap. As more ants follow the path with the highest pheromone concentration the pheromone in the other paths begin to evaporate with time. Thus, they compute the optimal path.
In this paper we applied the traditional ACO algorithm to solve an LPP in the continuous domain by thorough the method of recursive discretization. We used the constraints of the problem to define the boundaries of our solution. The algorithm is explained below.
A. Algorithm
1. Create Initial population of ants;
2. Set the boundaries of space for search using
constraint equations.
3. Discretize continuous domain into clustered points.
At first large size clusters are formed.
4. Spawn the ants at random location in space.
5. While(discretization factor > Set Value)
6. for i=1:n (all n ants)
7. Calculate the desirability of ants next location using
cost function and pheromone presence.
8. Move ant to most desirable point and increment the
pheromone value of that point.
9. end if
10. Ants cannot move from presents location.
11. Choose point with least cost value.
12. Define new space.
13. Decrement discretization.
14. end
First a fixed population of ants are spawned in a space defined by the constraints. The ants begin foraging by moving from one point to another by determining the desirability of the point and choosing the point with maximum desirability. The desirability of a point is determined by the function
Pm,n = (ταm,n)*(ηβ
m,n) (9)
Σ(ταm,n)*(ηβ
m,n)
Where m,n are the x and y coordinates of a point.
τ amount of pheromone on that point and α is the factor which controls the influence of pheromone.
η desirability of point based on cost function defined by the equation to be maximized and β is the factor which controls the influence of the cost function.
After moving to a point the ant updates the pheromone value of the point using the equation.
τmn = (1-ρ)τmn + ΣΔτkmn (10)
where
ρ pheromone evaporation rate
τkmn amount of pheromone deposited by the kth ant.
Δτkmn is calculated by the formulae,
Δτkmn = Q (11)
Lk
where
Q constant factor
Lk cost of the kth ant’s tour which in this case is the value of the cost function.
At the end of each iteration the minimum points of
the function are determined and the point which gives the least
value is defined as the new space for next iteration with an
decrease in the discretization factor.
V. RESULTS
2-bus Radial system
27
Consider the 2-bus radial system shown in Figure 1. It includes a 220 kV, 100 MVA source (also taken as the base kV and base MVA of the system). The CTI for the relay is taken as 0.57 s. The maximum fault current just beyond relay R1 is 2108A and beyond R2 is found to be 1703 A. Using the equations (2) and (4) the values of α are calculated and tabulated as shown in Table 1. Here we assume the upper limit of the TMS of both relays as 1.2 and the lower
R2 being the primary relay operates first when the fault occurs at F. Let R2 operate 0.2 s after the fault inception to ensure that it does not operate for current surges. Relay R1 should operate after the CTI, which equals to the sum of operating time of circuit breaker (CB) at bus 2, overshoot time of relay R1 and 0.2 sec.
TABLE 1
Relay Constants
FAULT LOCATION RELAY RA- Ap
CONSTANTS RELAY RB- Ap
CONSTANTS
JUST BEYOND A 3.21
JUST BEYOND B 7.38 3.57
Let x1 and x2 be TMS values of relay R1 and R2 respectively.
3.57*x1 – 3.57*x2 ≥ 0.57 (12)
Subject to the constraint,
3.21*x1 ≥ 0.2 (13)
And
3.57*x1≥ 0.2 (14)
The upper limit is taken at 1.2.
For (ACO) we set the discretization factor to 1 at first
and then decrement by one tenth of the original value for each
iteration for 5 such iteration to get our value our minimum
point to an accuracy of 10-5. A population is chosen of 50 ants
to start with as this helps to arrive at a minimum point faster.
The results of ACO are compared with the Two Phase
Simplex Algorithm and are tabulated as shown in Table 2.
Table 2 TMS values of relays
TMS of Relays
X1 X2
Two
Phase
simple
x
.215 .056
ACO 0.11574 0.05851
VI. INFERENCE
To optimize the TMS, two algorithms namely Ant
Colony Optimization and Two Phase simplex method have
been compared. It can been deduced that using ACO, the
optimization is higher and more effective
VII. CONCLUSION
The protection of distribution system from
overcurrent faults is very important for power system
protection engineers. OC relays are predominantly used in
radial distribution networks and are expected to identify and
isolate faults instantly. This paper compares the effectiveness
of Swarm Intelligence Algorithm ACO and Two Phase
Simplex Algorithm in finding the optimized solutions for time
multiplier setting and time of operation of relays. These
algorithms can also be conveniently extended to larger
distribution networks. Moreover, to better understand
distribution networks and to perceive its ability to minimize
and isolate faults, other optimization can be undertaken to
obtain case specific results.
REFERENCES
1. O.V.G. Swathika, and S. Hemamalini, “Prims-Aided Dijkstra Algorithm for
Adaptive Protection in Microgrids,” IEEE Journal of Emerging and Selected
Topics in Power Electronics, vol. 4(4), pp.1279-1286, 2016.
2. O.V.G.Swathika, A. Das, Y. Gupta, S. Mukhopadhyay, and S. Hemamalini,
“Optimization of Overcurrent Relays in Microgrid Using Interior Point
Method and Active Set Method,” In Springer Proceedings of the 5th
International Conference on Frontiers in Intelligent Computing: Theory and
Applications, pp. 89-97, 2017.
3. A. Gupta, O.V.G. Swathika, and S. Hemamalini, “Optimum Coordination
of Overcurrent Relays in Distribution Systems Using Big-M and Dual
Simplex Methods,” In IEEE Computational Intelligence and Communication
Networks, pp. 1540-1543, 2015.
4. O.V.G. Swathika, S. Mukhopadhyay, Y. Gupta, A. Das, and S.
Hemamalini, “Modified Cuckoo Search Algorithm for Fittest Relay
Identification in Microgrid,” In Springer Proceedings of the 5th International
Conference on Frontiers in Intelligent Computing: Theory and Applications,
pp. 81-87, 2017.
5. O.V.G. Swathika, S. Hemamalini, “Review on Microgrid and its Protection
Strategies,” International Journal of Renewable Energy Research, vol. 6(4),
pp.1574-1587, 2016.
6. O.V. Gnana Swathika, Indranil Bose, Bhaskar Roy, Suhit Kodgule, and S.
Hemamalini, “Optimization Techniques Based Adaptive Overcurrent
Protection in Microgrids,” Journal of Electrical Systems, Special Issue 3, vol.
10, 2016.
7. O.V. Gnana Swathika, and S. Hemamalini. Adaptive and Intelligent
Controller for Protection in Radial Distribution System. In Springer Advanced
28
Computer and Communication Engineering Technology, vol. 362, pp. 195-
209, 2016.
8. M. Sukumar Brahma and A. Adly Girgis, “Development of Adaptive
Protection for Distribution Systems With High Penetration of Distributed
Generation,” IEEE Transactions on Power Delivery, vol.19, pp. 56-63, 2004.
9. C. Vassilis Nikolaidis, Evangelos Papanikolaou, and S. Anastasia
Safigianni. “A Communication-Assisted Overcurrent Protection Scheme for
Radial Distribution Systems with Distributed Generation,” IEEE Transaction
on Smart Grid, 7(1), 114-123, 2016.
10. H.H. Zeineldin HH, E. F. El-Saadany, and M.M.A. Salama. “Distributed
Generation Micro-Grid Operation: Control and Protection,” In Power Systems
Conference: Advanced Metering, Protection, Control, Communication, and
Distributed Resources, pp. 105-111, 2006.
11. P. Prashant Bedekar, R. Sudhir Bhide, and S. Vijay Kale, “Optimum
Coordiantion of Overcurrent relays in Distribution system using Dual Simplex
Method”, In IEEE International Conference on Emerging Trends in
Engineering and Technology, pp. 555-559, 2009.
12. Y.G. Paithankar, and S.R. Bhide, “Fundamentals of Power System
Protection,” Prentice Hall of India Private Limited, New Delhi, 2007.
13. Badri Ram, and D.N. Vishwakarma, “Power System Protection and
Switchgear,” Tata McGraw Hill Publishing Company Limited, New Delhi,
2008.
14. B. K. Manohar Singh, B. K. Panigrahi and A. R. Abhyankar, “Optimal
Overcurrent Relay Coordination in Distribution System”, In IEEE
International Conference on Energy, Automation, and Signal, pp. 1-6, 2011
15. K. Deb.”Optimization for Engineering Design –Algorithms and
Examples,” Prentice Hall of India Private Limited, New Delhi, 2006.
16. P.P. Bedekar, and S.R. Bhide, “Optimization of multivariable nonlinear
functions using genetic algorithms,” In IEEE International Advance
Computing Conference, 2009.
17. R. Madhumitha, P. Sharma, D. Mewara, O.V.G. Swathika, and S.
Hemamalini, “Optimum Coordination of Overcurrent Relays Using Dual
Simplex and Genetic Algorithms”, In IEEE International Conference on
Computational Intelligence and Communication Networks, pp. 1544-1547,
2015.
18. Chao-Rong Chen, Cheng-Hung Lee, and Chi-Juin Chang, “Optimal
Overcurrent Relay Coordination in Power Distribution System Using a New
Approach”, International Journal of Electrical Power & Energy Systems,
45(1), pp.217-222, 2012.
29
An Improved Rules-based Control of Battery Energy
Storage for Hourly Power Dispatching of
Photovoltaic Sources
M. A. Jusoh and M. Z. Daud School of Ocean Engineering, Universiti Malaysia Terengganu,
21030 Kuala Nerus, Terengganu, Malaysia.
Abstract—Battery energy storage (BES) system is effective in
smoothing and dispatching the fluctuation output from solar
photovoltaic (PV) sources. This paper presents an improved rules-
based control scheme for BES with the goal of minimizing the
fluctuation output from PV sources while ensuring the operational
constraints of BES are regulated at the specified ranges for the
safety purposes. The control scheme is developed based on the
desired operational constraints of BES such as state-of-charge
(SOC) and charge/discharge current limits. The simulation studies
were carried out by using Matlab/Simulink to evaluate the
effectiveness of the proposed control scheme on the 1.2 MW PV
system data obtained from a site in Malaysia. Furthermore, a
comparative study of the proposed control scheme with the
existing methods has been done to address the effectiveness of the
control scheme. The simulation results show that the proposed
control scheme can effectively minimized the output power
fluctuations of the PV sources and dispatching the output on an
hourly basis to the utility grid with the efficiency up to 94.47%.
Finally, the comparison results also illustrates the proposed
control scheme as the most effective controller compared to the
other type of controllers studied previously.
Index Terms—Power fluctuation mitigation; Battery energy
storage; Lithium-ion battery; Photovoltaic system.
I. INTRODUCTION
Solar photovoltaic (PV) energy source is well known for its
unpredictable and inconsistent output due to the intermittent
nature of solar irradiance and temperature [1]. High penetration
of unpredictable and inconsistence of solar PV output into
utility grid system caused many problems including voltage and
power fluctuations and other power quality problems.
Integration of solar PV system with battery energy storage
(BES) system is proven to be effective on minimizing such
problems provided that a proper control scheme is designed and
managed [1-4]. Various types of batteries are potential to be
integrated to solar PV system for power fluctuation mitigation
purposes such as Lead acid (LA), Lithium-ion (Li-ion) and
Nickel Cadmium batteries [5]. However, high cost of BES
system is considered one of the obstacles that require further
attention. For many cases, studies associated to developing a
robust and efficient control method for BES are of significant
importance to provide a cost-effective BES system.
There are many types of control schemes for BES system
were proposed in the literature for the purpose of smoothing
fluctuation output of renewable energy sources. However, few
researches have been focusing on the smoothing fluctuation
output with constant output power dispatching. In [2], the
optimization-based state-of-charge feedback (SOC-FB) control
scheme for the valve-regulated lead acid (VRLA) BES has been
proposed to regulate the SOC of BES according to the desired
operational constraints such as SOC and current limits during
the smoothing and dispatching processes. The authors proposed
genetic algorithm-based parametric optimization with overall
smoothing and hourly dispatch efficiency recorded equal to
84%. Consequently, heuristic optimization-based studies have
been investigated in [3] using other algorithms such as
gravitational search algorithm (GSA), and particle swarm
optimization (PSO). For this case, the results for the fluctuation
mitigation efficiency was measured 89.91% using GSA
approach. However, long computation times and accurate BES
system model were required for the optimization processes. In
[4], a simple approach based on the rules has been proposed.
The rules in the control scheme was determined based on the
desired operational constraint of BES system. The results
showed the proposed rules-based control scheme effectively
smoothing the fluctuation output of PV system. However, it was
failed to sustain the BES power at the desired level. In this
regards, to overcome the associated issues in [2-4], this paper
introduces an improved rules-based control scheme for BES
system. The objective of the study is to develop a simple and
robust control scheme for BES system so that the fluctuation
output can be smooth out and dispatch out to utility grid system.
The rest of the paper is organized as follows: Section II
describes the details of the proposed control scheme and the
simulation set-up. Section III presents the results and discussion
from the simulations. Finally, Section IV concludes the paper.
II. METHODOLOGY
In the present study, a typical AC-coupled PV-BES structure
for power smoothing and power dispatch is presented in Figure
1. The PV and BES systems are parallel connected to the PCC
via bi-directional voltage-sourced-converter, VSC (i.e. PV-
VSC and BES-VSC). The BES-VSC system is responsible in
regulating the fluctuated output of PV system (PPV) through
charge and discharge of BES power (PBES). In order to provide
safety and economical operation of BES system, the BES
operation is subjected to several desired operational constraints
as described in equation (1)-(3), where SOCBES,min and
30
SOCBES,max are the minimum and maximum level of SOC
operating ranges of BES (SOCBES). The IBES,min and IBES,max are
the minimum and maximum allowed current of BES (IBES) and
VBES,min and VBES,max are the operational constraints of BES
voltage (VBES). The SOCBES,min and SOCBES,max are set to 0.3 p.u
and 0.9 p.u, respectively, which is 60% of the total capacity of
BES. The SOCBES constraint is used to prevent the BES from
the over-discharge and over-charge. The IBES constraint is set
to ±1×CBES based on the BES-VSC current limit, while the VBES
constraint is set to 10% of the BES rated voltage. The VBES
operational is used to prevent the BES from the breakdown.
max,min, )( BESBESBES SOCtSOCSOC (1)
max,min, )( BESBESBES ItII (2)
max,min, )( BESBESBES VtVV (3)
Figure 1: General structure of PV-BES system
The BES-VSC employed the current-mode control strategy
that has two loops as presented in Figure 1. The details of
current-mode control are discussed in [6]. The outer control
loop is used to generate the reference current (IBES,ref) signal
either to charge or discharge while inner control loop is used to
generate the switching signals for the BES-VSC. In order to
generate the optimal IBES,ref for output power smoothing and PV
power dispatch, control scheme is introduced in outer control
loop as discussed in the following section.
A. Development of Control schemes for BES system
i. PSO-based SOC-FB control scheme
This optimization-based SOC-FB control scheme of BES
system has been proposed in [2] for hourly power dispatch of
PV system. In this paper, the optimization-based control
scheme is developed for the purpose of comparison of the
control scheme performances. Figure 2 illustrates the overall
diagram of optimization-based SOC-FB control scheme. The
aim of the control scheme is to minimize the deviation between
hourly forecasted power reference, PSET and PPV and generate
the optimal IBES,ref for BES system while ensuring the
operational constraints of BES are regulated at the specified
ranges. As illustrates in the Figure 2, PSO algorithm is used to
find the optimal parameters of SOC-FB control scheme and the
capacity of the BES system. The objective function of the
optimization is determined based on the Equation (4), where the
vector x represents the SOC-FB control scheme parameters
(MSOC, TSOC) and the capacity of BES system (CBES).
( ) −= dt(t)P(t)POF(x) GSET2
min (4)
Figure 2: Optimization-based control scheme
ii. Improved-rules based control scheme
The conventional rules-based control scheme of BES system
for hourly dispatch of PV system output has been introduced in
[4] with the same objective as the optimization-based SOC-FB
control scheme. The conventional rules-based is simple, require
minimal computation times and does not required accurate BES
system model compared to optimization-based control scheme.
However, the conventional rules-based control scheme can only
control the BES from over-charge and over-discharge but not
able to sustain the BES power at the desired level. Such an
imperfection makes the conventional controller unable to
support the continuous BES power required for dispatching
operation of intermittent PV output power.
The rules of the controller are divided into two parts: i.e. rules
for SOCBES constraint, and IBES constraints, respectively. For
SOCBES constraint, the rules are created to guarantee the SOCBES
to be kept within the desired limit (SOCBES,min and SOCBES,max)
during the smoothing and dispatching process. Meanwhile, in
the IBES constraints, the rules are developed to limit the charging
and discharging current of BES at the desired operational limit
(IBES,min and IBES,max). In the present work, some improvements
to the conventional rules-based control scheme have been
suggested and developed. Overall structure of the improved
DC
DC
DC
AC
DC
AC
PV System PV-VSC
BES-VSC DC Bus
PCC
Grid system
Lf 2
Lf 1
PPV
PBES
PG
CB
PV-conv
with MPPT
control
BES
Switching signal
Inner control loop
Power smoothing and hourly power
dispatch control scheme in outer
control loop
Current-mode Control scheme
PSET
’
PPV
31
control scheme is illustrates in the Figure 3. As illustrates in the
figure, the SOC power correction of (PSOC) is added to the
conventional rules-based control scheme to ensure the SOCBES
is maintain at the desired SOC level (SOCBES,ref) from the
beginning of the process until the end. The PSOC in unit of MW
is applied to PBES,tar signal, where positive and negative value
of PSOC represents the shortage and the surplus energy of BES
to maintain SOCBES at SOCBES,ref, respectively.
Figure 3: Improved rules-based control scheme
B. Simulation set-up and evaluation of the performance of
the control methods
The simulation study is carried out using Matlab software.
The PPV and PSET data are obtained from the daily average of
1.2 MW PV system output measured in Malaysia [2].
Meanwhile, Li-ion-types of PowerSim battery model in
Matlab/Simulink is used as BES energy storage. For the
optimization-based control scheme, the SOC-FB control
scheme is developed in Matlab/Simulink, while the PSO
algorithm for the optimization process is developed in M-file.
During the simulation process, the Matlab/Simulink is linked to
PSO algorithm in M-file and the processes running
simultaneously. The simulation process is terminated when the
PSO algorithm meet the optimal values at the optimal
condition. Meanwhile, for the improved rules-based control
scheme, the rules in the control scheme is implemented in
Matlab/Simulink using Matlab function block, where the
C/C++ code language is used to represent the rules.
In order to evaluate the robustness and flexibility of the
performance of the control schemes, the case studies of the
varying BES capacity are considered. The purpose of the BES
capacity case studies is to verify the effect of the BES capacity
to the control scheme performance. For this cases, the capacity,
CBES, is set to 0.25 MWh (416.7 Ah), 0.3 MWh (500 Ah) and
0.35 MWh (583.3 Ah), respectively. In addition, analysis of
simulation results is also given through evaluation using the
Performance Index (PI) [4], Battery Health Index (BHI) [7] and
Efficiency (ɳ) [2] as given in equations (5)-(7). The value of Nx
of equation (5) represents the number of occurrence of the
deviations, while dPx represents the deviation between, powers
delivered to grid, PG, with respect to power reference set-point,
PSET. The PI0 and PIBES of equation (7) represent performance
index without/with using BES, respectively. For dP criteria, it
is assumed that the deviations up to ± 0.1 MW are acceptable.
Meanwhile, for PI and BHI, a smaller values indicates the high
dispatching performance and the effective usage of BES device
within the safe operating limits.
𝑃𝐼 = ∑ 𝑁𝑥 × |𝑑𝑃𝑥| (5)
=
−=
T
t
refBESBES SOCtSOCT
BHI
1
2,)(
1 (6)
( ) 100/ 00 −= PIPIPI BES (7)
III. RESULTS AND DISCUSSION
The results are divided into two parts. The first part discusses
about the effects of improved control scheme to the dispatching
performance of the PV-BES system, whereas the second part
provides the effects of the BES capacity to the improved control
scheme performance and the PV-BES system dispatching
performance compared to other control schemes.
A. Dispatching performance of PV-BES system by using
improved rules-based control scheme.
Firstly, for the preliminary study, the simulation is carried out
to present the dispatching performance of PV-BES system
without control scheme. This case considers BES size of 0.3
MWh with initial capacity set at 60% of the total capacity. In
addition, it is assumed that the BES system will be disconnected
from the PV system if the SOCBES is outside the desired
operating constraints of BES. Figure 4 shows the results of the
dispatching performance of PV-BES system without control
scheme of BES.
Figure 4: Dispatching performance of PV-BES system without control scheme
From figure 4(a) and 4(b), it is evident that the difference in
accuracy of forecasted power set-point, PSET, affects the
dispatching performance of the PV-BES system without control
schemes. For the case of 100% accuracy of PSET, the output
power fluctuation of PPV is successfully smoothed and
dispatched to the grid while meeting all the desired operating
constraints until 7 PM in the afternoon. However, at 90%
accuracy of PSET, the output power fluctuation of PPV is
7 9 11 13 15 17 19
(a)
Po
wer
(M
W)
0.0
0.4
0.8
1.2
PG (100% accuracy of P
SET)
Time (hr)
7 9 11 13 15 17 19
(b)
SO
C (
p.u
)
0.0
0.2
0.4
0.6
0.8
1.0
SOCBES, max
SOCBES, min
with BES without BES
PG (90% accuracy of P
SET)
32
successfully smoothed and dispatched only from 7 AM to 8 AM
due to the disconnected BES system from the system. The BES
is disconnected from the system due to the SOCBES level that
has reached the SOCBES,min as illustrates in Figure 4(b). From the
results, it can be concluded that without control scheme applied
to the system, larger BES capacity is desired in order to keep
the BES continuously operated within the SOCBES level.
Therefore, to meet the acceptable dispatching performance
using minimum capacity of BES, the SOCBES needs to be
properly controlled.
Figure 5(a)-(d) presents the dispatching performance of PV-
BES system output power, VBES profiles, SOCBES profiles and
IBES profile at 90% accuracy of PSET using 3 different control
schemes, respectively. In this case, an optimal 0.287 MWh (478
Ah) of CBES that has been determined by using optimization-
based control is used. As shown in Figure 5(a), the dispatching
performance of PV-BES system by using improved rules-based
control scheme and optimization-based control scheme can
track the PSET perfectly while keeping the BES operational
constraints at the desired limits. As evident from the Figure
5(b)-(d), all operating constraints of BES are varied within the
desired operating ranges. There are some spikes exist in the
dispatching output mostly between 11 AM and 3 PM because
of the current block inside the controller for safe operation
purposes. Meanwhile, Figure 5(a) also illustrates the poor
dispatching performance of conventional rules-based control
scheme. The conventional rules-based nearly failed to track the
PSET perfectly in between 1 PM to 7 PM due to the insufficient
of BES energy.
Figure 5: Dispatching performance of PV-BES using different control
schemes
Figure 6(a)-(d) illustrates the histograms and the normal
distribution curves of dP for comparing the performances
without deployment of control scheme and with using 3
different control schemes, respectively. From normal
distribution in Figure 6(a)-(d), the percentage of occurrences of
unacceptable deviations are obtained. For the case of without
control scheme, percentage of occurrences of unacceptable
deviations is calculated equal to 42.1%, while for optimization-
based (SOC-FB), conventional rules-based and improved rules-
based control schemes are 0.08%, 3.64% and 0.01%,
respectively. The obtained results prove that, the performance
of the improved rules-based control scheme is better than
optimization-based SOC-FB, and conventional rules-based
control scheme, respectively.
Figure 6: Histogram and normal distribution of dP
Finally, Table 1 gives the details of extracted results from the
Figure 6(a)-(d). From the results, it clearly shows that the
improved rules-based control scheme is more efficient in
dispatching process with the performance index, PI, of 36.83
and efficiency of 94.5% compared to the other control schemes.
Besides that, in terms regulation of the state-of-charge, SOCBES,
using improved control scheme, the SOCBES is maintained at
minimum SOCBES of 0.43 p.u and the BHI measured around
0.0817. From the results, it can be concluded that the improved
control scheme can optimally smoothing the fluctuation of PPV,
and can extend the lifetime the of BES system.
B. Effect of capacity, CBES, to the dispatching performance
of PV-BES system
Table 2 presents the effect of BES sizes on the dispatching
performance for each control scheme, respectively. For the
7 9 11 13 15 17 19
(a)
Po
wer
(M
W)
0.0
0.4
0.8
1.2
PG of Improved rules-based
7 9 11 13 15 17 19
(b)
Vo
ltag
e (k
V)
0.60
0.64
0.68
7 9 11 13 15 17 19
(c)
SO
C (
p.u
)
0.0
0.5
1.0
Time (hr)
7 9 11 13 15 17 19
(d)
Curr
ent
(kA
)
-1.0
-0.5
0.0
0.5
1.0
PG of Optimization-based
PG of Conv. rules-based
0.35
0.16
0.00
-0.1
6
-0.3
5
12
9
6
3
0
dP (MW)
Occ
ura
nce
(%
)
0.14
0 .07
0 .00
-0.0
7
-0.1
4
24
18
12
6
0
dP (MW)
Occ
ura
nce
(%
)
0 .18
0.12
0 .06
0.00
-0.06
-0.1
2-0
.18
80
60
40
20
0
dP (MW)
Occ
ura
nce
(%
)
0.14
0.07
0.00
-0.0
7- 0
.14
30
20
10
0
dP (MW)
Occ
ura
nce
(%
)
(a) Without control scheme (b) SOC-FB Control
(c) Conv. Rules-based control (d) Improved Rules-based control
Mean=0.01962 Mean=0.01803
Mean=0.00978 Mean=0.01931
SD=0.12265 SD=0.02239
SD=0.04425 SD=0.01897
33
unacceptable of dP, the results are evaluated and analyzed
through the normal distribution of dP. For optimization-based
control scheme, the unacceptable of dP is reduced from 0.08%
to 0.01% if the size of BES changed from 0.287 MWh to 0.35
MWh, while for conventional rules-based, the results show
reduction from 3.64% to 2.81%. However, for the improved
control scheme, the unacceptable of dP can be reduced further
to nearly 0%. Based on the results, it can be concluded that by
using improved control scheme, the size of the BES can be
reduced that contributes minimum cost.
Table 1
Results of dispatching performance of PV-BES using different control scheme
(BES=0.287 MWh)
Parameters (A)
Control Methods
Optimization-
based SOC-FB
Rules-based
Conventional Improved
PV capacity (MW) 1.2
BES capacity, CBES
(MWh) 0.287 (478 Ah)
Initial state-of-charge,
SOCi (p.u) 0.6
Terminal
voltage, VBES
(kV)
Max 0.6580 0.6512 0.6593
Min 0.6268 0.6172 0.6285
State-of
charge,
SOCBES (p.u)
Max 0.6000 0.6000 0.6274
Min 0.4033 0.3009 0.4337
Current, IBES (kA) ± 0.478
Performance index,
PI 43.5681 109.0115 36.8371
Battery health
index, BHI 0.0996 0.2525 0.0817
Efficiency, (%) 93.4634 83.6447 94.4732
Table 2
Effects of BES size to the dispatching performance of PV-BES system
Parameters
(unit) BES size
(MWh)
Control Schemes
Optimization-
based SOC-FB
Rules-based
Conventional Improved
Unacceptable
of dP
(%)
0.25 9.58 5.49 0.09
0.287 0.08 3.64 0.01
0.35 0.01 2.81 0
Performance
index, PI
0.25 155.7277 117.2707 45.2869
0.287 43.5681 109.0115 36.8371
0.35 41.0042 98.7146 30.5195
Battery
health index,
BHI
0.25 0.2180 0.2494 0.0923
0.287 0.0996 0.2525 0.0817
0.35 0.0816 0.2539 0.0681
Efficiency,
(%)
0.25 76.6358 82.4056 93.2055
0.287 93.4634 83.6447 94.4732
0.35 93.8480 85.1896 95.4211
In terms of the PI, BHI and efficiency, the results are also
provided in Table 2, respectively. From the PI results, by using
0.35 MWh BES, the PI of improved rules-based can be
achieved up to 30.5. However, for optimization-based and
conventional rules-based, the PI are only reduced up to 41.0 and
98.7, respectively. In terms of the effects of the BES size to the
BES BHI, increased BES size can decrease or in other words,
improvise the BHI for the case using optimization-based and
improved rules-based control scheme, respectively. On the
other hand, the results are totally different to the case of
conventional rules-based controller. The decreasing BHI of the
former controller is due to reduction of the charging and
discharging depth level of BES. By using 0.35 MWh BES size,
the BHI for improved rules-based control scheme is 0.0681,
while for optimization-based and conventional are only 0.0816
and 0.2539, respectively. From the results, it is evident that the
increasing of BES size and the use of improved rules-based
controller can increase the lifetime of the BES. Finally, Table 2
also compares the efficiency of the control schemes. Based on
the results, the improved rules-based control scheme performed
at the highest efficiency compared to the other control schemes.
For example, using 0.35 MWh BES size, the efficiency of
improved rules-based control scheme is 95.4%, compared to the
efficiency of optimization-based and conventional control
scheme at 93.8% and 85.2%, respectively. Overall of the results
show 10.2% improvement of efficiency of the improved rules-
based control scheme compared to the conventional rules-based
controller.
IV. CONCLUSION
An improved rules-based control scheme for BES system to
smooth out and dispatch the fluctuation of PV system on an
hourly basis to the utility grid is presented. The operational
constraints for BES are considered to ensure safe operation of
the system whilst providing power regulation service
continuously. Simulation results shows good performance of
the proposed control scheme compared to other previously
developed methods through the analysis of efficiency that has
been measured around 94.47% (at BES = 0.287 MWh). The
results also show that only proposed control scheme can reduce
the unacceptable deviation completely by using 0.35 MWh BES
system. The BHI also the lowest: i.e. 0.0817 (at BES = 0.287
MWh) compared to other control schemes. The overall results
clearly indicate the capability of the proposed control scheme
in increasing the lifetime of the BES system when it is subjected
to continuous charge/discharge operations particularly in power
fluctuation mitigation of solar PV sources.
ACKNOWLEDGMENT
This work is supported by Ministry of Higher Education
Malaysia (MOHE), Malaysia under the Fundamental Research
Grant Scheme (FRGS), Vot No. 59418 (Ref:
FRGS/1/2015/TK10/UMT/02/1).
REFERENCES
[1] S. Shivashankar, S. Mekhilef, H. Mokhlis, and M. Karimi, "Mitigating
methods of power fluctuation of photovoltaic (PV) sources–A review,"
Renew. Sust. Ener. Rev., vol. 59, pp. 1170-1184, 2016.
[2] M. Z. Daud, A. Mohamed, and M. Hannan, "An improved control method of battery energy storage system for hourly dispatch of photovoltaic power
sources," Energ. Convers. Manage., vol. 73, pp. 256-270, 2013.
[3] M. Z. Daud, A. Mohamed, A. A. Ibrahim, and M. Hannan, "Heuristic optimization of state-of-charge feedback controller parameters for output
power dispatch of hybrid photovoltaic/battery energy storage system,"
Measurement, vol. 49, pp. 15-25, 2014.
34
[4] S. Teleke, M. E. Baran, S. Bhattacharya, and A. Q. Huang, "Rule-based
control of battery energy storage for dispatching intermittent renewable
sources," IEEE Trans Sustain Energy, vol. 1, pp. 117-124, 2010.
[5] M. Delfanti, D. Falabretti, and M. Merlo, "Energy storage for PV power plant dispatching," Renew. Energ., vol. 80, pp. 61-72, 2015.
[6] T. Suntio, T. Messo, and J. Puukko, Power Electronic Converters:
Dynamics and Control in Conventional and Renewable Energy Applications: John Wiley & Sons, 2017.
[7] T. T. Trung, S.-J. Ahn, J.-H. Choi, S.-I. Go, and S.-R. Nam, "Real-time
wavelet-based coordinated control of hybrid energy storage systems for denoising and flattening wind power output," Energies, vol. 7, pp. 6620-
6644, 2014.
35
Analysis of the Coexistence of Wi-Fi Networks and
Real-Time Positioning Systems for Data Trafficking
and Object Positioning.
G. Cuzco1, 2, H. Moreno2 1Universidad Nacional de Chimborazo.
2Escuela Superior Politécnica de Chimborazo.
Abstract—Due to the increasing behavior of connected devices,
the development of technologies such as wearables, smart cities,
and others, announce a saturation of WiFi networks at 2.4HGz,
reducing their performance and encouraging migration at 5GHz.
This paper describes the implementation of a real-time
localization system with IEEE standard 802.15.4-2011 in order to
determine the coexistence between WiFi networks operating in
the 5GHz band and real-time location systems. An experimental
study has been conducted in open and closed environments; the
location system has been implemented using the DW1000
tranceiver with UWB antenna. Traffic, speed and Cartesian
coordinate’s data were collected in each scenario, traffic and
speed data were collected with Wireshark and Colasoft Capsa
software packages, Cartesian coordinates were collected in
Matlab and in conjunction with the rest of the information
analyzed in SPSS. Being the purpose to analyze the coexistence
between WiFi systems and location systems, it has been possible
to determine the affectation in the dispersion of the data in the
location systems in real time. The results obtained will be useful
when planning electronic systems that use these technologies so it
is recommended to analyze the transmission channels trying to
keep them with conservative guard bands and the need to
improve the algorithms for data management in location systems.
Index Terms— Coexistence of Protocols; IEEE 802.11a; IEEE
802.15; Location Systems.
I. INTRODUCTION
Since its inception, wireless communications have played a
preponderant role to keep us connected everywhere,
depending on WiFi connections to maintain our productivity
standard, thus promoting the development of dedicated
technologies in the area of health, security, home automation
and the leverage of the expression of the internet of things
transforming from a future trend to a reality.
The location systems in outdoor areas are covered by global
positioning systems that we know as Gps [18] that are
responsible for measuring the arrival time of radio signals and
with this information, calculate their position.
Indoors, it is using triangulation methods and algorithms
based on received signal strength or arrival time, which are
used to improve performance by using Ultra Wide Band
(UBW) based transmission systems governed by the IEEE
802.15-2011 standard. Therefore, the analysis of the influence
of wireless channels that operate in the WI-FI bands with the
802.11a standard on the accuracy of a positioning system
which is implemented by triangulation will be considered as a
research problem of radio frequency signal level and other
methods using the Ultra-Wide band technique.
II. METHODOLOGY
A scheme has been used where a communication system
based on two computers operates that uses the IEE 802.11a
standard. In the same way, a real time localization system
(RTLS) has been implemented, which operates with the IEEE
802.15.4-2011 standard.
The systems will always operate under the same specific
conditions controlling the variable of separation between them
and reporting the variables that correspond to each system
which will be detailed later, the general scheme presented can
be seen in figure 1. The tests have been developed in two
scenarios, one in open environments and another scenario
called closed environments.
Figure 1: General operation scheme
A. Communication System
In the antennas used, the transmission channel has been
configured with Access Point mode with a center frequency of
5840MHz with a channel width of 40MHz
.
36
Figure 2: Wireless transmission scheme
SCENARIO A. Closed Environments
The wireless communication system is made up of two
antennas of the Ubiquiti brand, which are separated by a
distance of 6.5m in an office environment. For the analysis of
network traffic, software packages are used: Wireshark and
Colasoft Capsa, which allow monitoring and analyzing
communication networks. The values obtained by each
software are analyzed by IBM SPSS Statistics.
These values will be obtained in three cases: a) Ubiquiti
antenna at 3m from the RTLS system, b) Ubiquiti antenna at
6m from the RTLS system and c) RTLS system deactivated.
SCENARIO B. Open Environments
The wireless communication system is made up of two
antennas of the Ubiquiti brand, which are separated by a
distance of 30m in an open environment, the same ones that
were installed on concrete poles in the sector. For the analysis
of network traffic, it uses software packages: Wireshark and
Colasoft Capsa, which allow monitoring and analyzing
communication networks. The values obtained by each
software are analyzed by IBM SPSS Statistics.
These values will be obtained in three cases: a) Ubiquiti
antenna at 3m from the RTLS system, b) Ubiquiti antenna at
6m from the RTLS system and c) RTLS system deactivated.
B. Real Time Positioning System (RTLS)
The real-time positioning system is composed of three
devices called fixed anchors and an element of similar
characteristics called Tag, the system used corresponds to the
Decawave family. The operating model used is shown in
Figure 3.
Figure 3: RTLS model used
RTLS anchors
The cards assigned as anchors have been installed at the
same height, the differences in mounting height influence in
the reduction of the location accuracy, in the same way the
installation of the anchors with line of sight (LOS) between
the cards
The propagation and detection model has been selected in
which one of the anchors reports the information to the PC
and, the mobile TAG does not exert additional actions, in this
way the anchors provide the necessary information to estimate
the position of the device assigned as TAG .
Tranceiver Decawave DW1000
The function of the tranceptor used is constituted by a single
Ultra-Wide Band (UBW) chip that consists of a transmitter
and an analogue receiver that through a digital tranceiver,
through an SPI interface is linked to the external Host. An
internal switch switches the antenna port to the TX / RX
function, transmission / reception.
Figure 4: DW1000 block Diagram
UWB antenna
The RF connection to the outside of the tracker DW1000 is
made through a pair of 100Ω differential pins which is
designed to work at a frequency of 3 to 8 GHz with an omni-
directional radiation pattern, provides a gain of 2.2 dBi at 4
GHz and 3.3 dBi at 6.5 GHz as seen in Figure 5.
37
Figure 5: UWB Antenna
Figure 6: Radiation Patterns: a) Azimuth plane theta 900, b) Elevation phi
00.
Figure 7: Maximum Gain antenna
III. RESULTS
The results are classified in the following:
1) Results of network traffic analysis
2) Location accuracy results (coordinates)
For this, two scenarios have been chosen: scenario A, closed
environments and; Scenario B, open environments, in each
scenario the two tests are planned, the data transmission and
the location system.
A. Analysis of network traffic
These values will be obtained in three cases:
a) Ubiquiti Antenna 3m from the RTLS system,
b) Ubiquiti Antenna 6m from the RTLS system and
c) RTLS system deactivated
SCENARIO A. Closed Environments
Table 1
Analysis of the network with the Colasoft software
Case Bytes
sent
(MB)
Packet
sent
Lost
Bytes
(MB)
Lost
Packet
Case
A
409,67 283,046 0,008 0,176
409,72 283,097 0,008 0,174
409,67 283,071 0,007 0,172
409,69 283,075 0,008 0,175
409,8 283,827 0,008 0,183
Case
B
409,67 283,082 0,007 0,17
409,67 283,07 0,007 0,169
409,67 283,072 0,007 0,169
409,67 283,097 0,007 0,17
409,7 283,675 0,007 0,169
Case
C
409,68 283,102 0,008 0,175
409,67 283,098 0,007 0,17
409,67 283,071 0,007 0,17
409,7 283,102 0,008 0,176
409,67 283,086 0,007 0,173
a)
c)
b)
38
SCENARIO B. Open Environments
Table 2
Analysis of the network with the Colasoft software
Caso
Bytes
sent
(MB)
Packet
sent
Lost
Bytes
(MB)
Lost
Packet
Case
A
409,69 283,075 0,008 0,175
409,7 283,102 0,008 0,176
409,67 283,097 0,007 0,17
409,72 283,097 0,008 0,174
409,67 283,046 0,008 0,176
Case
B
409,67 283,082 0,007 0,17
409,67 283,07 0,007 0,169
409,8 283,827 0,008 0,183
409,7 283,675 0,007 0,169
409,68 283,102 0,008 0,175
Case
C
409,67 283,072 0,007 0,169
409,67 283,071 0,007 0,172
409,67 283,086 0,007 0,173
409,67 283,071 0,007 0,17
409,67 283,098 0,007 0,17
In each case the trend of the behavior of the information has
been recorded as shown in table 3-1 and 3-2. The channel has
been verified its behavior with the spectrum analyzer as
indicated in figure 6.
Figure 6: Simultaneous behavior
A. Analysis of the position of objects
The real-time positioning system based on the DW1000
decawave tranceiver has been implemented in a closed
environment of 7m * 7m, for which five operating points have
been selected and the coordinates physically marked in both x
and y. The influence on the variable z is not analyzed in the
present project, so it has remained constant at 1m in height.
Table 3
Coordinates to perform the test
POINT X(m) Y(m)
P1 1 3.3
P2 2 2.5
P3 3.5 1.5
P4 4.5 0.7
P5 6.7 -0.4
The data collection is done at the distances established in
the initial parameters and its result is shown in Figure 7.
Figure 7: Fixed coordinates and coordinates with presence of interfering
signal (blue)
Similarly, the location of the antenna of the data link is
modified by moving it to a distance with less affectation, in
this case the distance to the nearest antenna is located at 6m
and the position data is recorded under these conditions.
Figure 8 shows the superimposed data of real position,
position without verified interference and position with the
operation in the conditions described above, the behavior of
the data is visible and is analyzed with statistical measures.
[m]
[m]
39
Figure 8: Fixed coordinates and coordinates with presence of interfering
signal distance two (blue)
As a result of the tabulation of the variations of the
dispersion of the measured data through the standard deviation
as a function of the distance, figure 9 is obtained, where it is
observed that the deviation decreases as the distance increases.
Figure 9: Behavior of the standard deviation
IV. CONCLUSIONS
In the present work an experimental procedure has been
developed to determine the affectation produced by the
wireless networks that operate at a frequency of 5.8GHz
causes the dispersion in the data collected by a location system
that uses the tranceiver DW1000 operating at close
frequencies.
The affectation found depends on the distance between the
two systems, which is more noticeable when they are carried
out in open environments, finding that at a greater distance,
the affectation is less.
In closed environments, the dispersion in the data is greater
in the positioning system even without the presence of
disturbances in the frequency band, this is due to the fact that
due to the method used for localization, it has effects on walls
and objects that change or delay the path of the signals.
. Acknowledgments The authors would like to thanks to the
Corporación Ecuatoriana para el Desarrollo de la
Investigación y la Academia -CEDIA-, for financing the
project “Control Coordinado Multi-Operador aplicado a un
robot Manipulador Aéreo”, – CEPRA-XI- 2017-06, for the
support to develop this paper.
REFERENCES
[1] Chávez García, P. (2012). Sistema de Localización en Interiores
por Ultrasonido. AVANCES EN CIENCIAS E INGENIERÍA.
[2] Ballazhi, R., & Farkas, D. (2012). Wireless Indoor Positioning Techniques. Communication Systems Seminar FS 2012.
[3] Carrera, E., & López Salomé. (20014). Event localization in
wireless sensor networks. PROCEEDINGS OF THE 2014 IEEE
CENTRAL AMERICA AND PANAMA CONVENTION.
[4] DecaWave. (2014). DecaWave Reference Planar Monopole
Antenna. DecaWave . [5] Decawave Limited. (2015). APS016: MOVING FROM TREK1000
TO A PRODUCT. Decawave Limited.
[6] DecaWave Limited. (2016). TREK1000 Expansion Options Instructions. DecaWave Limited.
[7] Decawave Limited. (2016). TREK1000 User Manual. Decawave
Limited. [8] DecaWave Ltd. (2015). TREK1000: Two-Way-Ranging (TWR)
RTLS IC Evaluation Kit. DecaWave Ltd.
[9] Decawave Ltd. (2016). DW1000 Device Driver API Guide. Decawave Ltd.
[10] DecaWave Ltd. (2014). TREK1000: Expansion Options.
DecaWave Ltd.
[11] Díaz, L., & Tabernilla, A. (s.f.). Sistemas de Localización en
interiores. Madrid: Universidad Politécnica de Madrid.
[12] Familia IEEE 802.11. (s.f.). [13] García Polo, E. (s.f.). Técnicas de Localización en Redes
Inalámbricas de Sensores. Universidad de Castilla.
[14] González, R., Dávila, N., & Lobalsamo, G. (s.f.). Pruebas de Alcance y Coexistencia de IEEE 802.15.4.
[15] IEEE Computer Society. (2011). Part 15.4: Low-Rate Wireless
Personal Area. IEEE Standards Assocciation. [16] Kanaris, L., Kokkinis, A., Liotta, A., Raspopoulos, M., & Stavrou,
S. (s.f.). A Binomial Distribution Approach for the Evaluation of
Indoor Positioning Systems. [17] Luděk Zaplatílek, M. P. (s.f.). Modeling of precise indoor
positioning system. Dept. of Electrotechnics, Faculty of Electrical
Engineering, University Pardubice.
[18] Marek Pola, L. Z. (2015). Measurement on precise indoor
positioning system in Real Enviroment. Dept. of Electrotechnics,
Faculty of Electrical Engineering, University Pardubice. [19] Mohammadreza , Y., & Bradford G., N. (s.f.). Obtenido de
http://www.cs.unb.ca
[20] Mok, E., Xia, L., Retscher, G., & Tian, H. (s.f.). A case study on the feasibility and performance of an UWB-AoA real time location
system for resources management of civil construction projects.
Journal of Applied Geodesy. [21] Popp, J., & Lopez, J. (2015). Real Time Digital Signal Strength
Tracking for RF Source Location. University of Washington, Seattle.
[22] Prieto Blázquez, J. (s.f.). Introducción a los sistemas de
comunicación inalámbricos. Universitat Oberta de Catalunya. [23] W.J. Lee, W. L. (s.f.). IEEE/ASME International Conference on
Advanced Intelligent Mechatronics. Design of Applications on
Ultra-Wideband Real-Time . [24] Xiong, Z., Song, Z., Scalera, A., Ferrera, E., Sottile, F., Brizzi, P., .
. . Spirito, M. (2013). Hybrid WSN and RFID indoor positioning
and tracking system. EURASIP Journal on Embedded Systems. [25] Zhao, L., Psota, E., & Pérez, L. (2014). A Comparison Between
UWB and TDOA Systems for Smart Space Localization. University
of Nebraska - Lincoln.
[m]
[m]
40
CHS Application on Novel Coplanar Routing
Azniza Abd Aziz, Chaitanya Sreerama
Abstract— Nowadays, the demands on higher bandwidth
computational performance has rapidly increased. At the same
time, increasing the platform and system size is not an option. This
brings new design challenges since the system bus will suffer signal
integrity degradation and limit performance. As system sizes
shrink, signal channels come closer together and crosstalk noise
will increase further limiting bus performance. In this paper we
will introduce a novel coplanar routing procedure based on the
application of “Crosstalk-Harnessed Signaling” (CHS) that will
achieve a system with higher bandwidth in small form-factor. The
CHS concept will help tackle the crosstalk problem and a Novel
Coplanar routing scheme will help increase the maximum bus
bandwidth per volume. CHS is one of the latest methods
addressing crosstalk issues that offers many advantages compared
to other existing methods. Novel Coplanar routing is an extension
of traditional microstrip routing that offers twice the routing
density compared to CHS microstrip routing. Besides that,
bandwidth per density will double if the Novel Coplanar routing
is applied to CHS 3D Novel routing.
Index Terms— Novel Coplanar routing; Crosstalk; Crosstalk
Harnessed Signaling; 3D Novel routing.
I. INTRODUCTION
In modern technology, the volume of smart phones and
tablets have increased rapidly reflecting higher demands on
internet usage in daily life. But the system needs to scale up and
there are design challenges to be overcome. Electrical signals
will suffer degradation as data rates increase in a densely routed
environment. One of the bottlenecks to system design
bandwidth is crosstalk noise. Thus, mitigating crosstalk helps
to improve the bandwidth demands.
There are multiple ways to mitigate the crosstalk such as
differential signaling, shielding or guard trace. These methods
are costly and inefficient for achieving high data rates
especially on high volume manufacturing(HWM). Besides that,
for differential signaling this requires twice signaling compared
to single ended and the guard trace method due to introducing
extra signals that will increase the board dimensions. Crosstalk
equalizer and eigen-mode signaling based on the modal
decomposition method is a new way to mitigate the
[2],[3],[4],[5]. These two concepts are promising for achieve
dense routing and higher bandwidth, but are very costly and
complex to design. In addition, this requires prior knowledge of
the channel characteristics and have complex termination
schemes.
Crosstalk Harnessed Signaling promises to be a method that
will improve on the modal decomposition and modal
composition schemes. This method was developed through
collaborative research between the University of South
Carolina and Intel. It offers lower cost, simple design, no
complex termination and no training is required on the channel.
This means that one static matrix works for different routing
types. Compared to other crosstalk mitigation schemes, a
different matrix is required for different routing schemes.
However, CHS shows it still works on CHS 3D Novel and
Novel Coplanar routing, which is introduced in this paper.
These two-routing schemes help to offer higher bandwidth per
volume.
II. CROSSTALK HARNESSED SIGNALING
Crosstalk Harnessed signaling (CHS) is a new technique for
mitigating crosstalk that offers better benefits compared to the
modal decomposition and modal composition methods. In this
paper, we will introduce the basic CHS method and then focus
on the CHS application in a new routing scheme. This technique
harnesses the crosstalk instead of eliminating it, thus
differentiating it from other crosstalk schemes. CHS offers
many benefits without requiring a complex transmitter (TX)
and receiver (RX) circuit design, nor complex termination. In
addition, no prior knowledge is required of the channel since
one CHS static matrix works for N channel lines.
Fig. 1 shows the CHS bus diagram [1],
Fig. 1 Crosstalk Harnessed Signaling (CHS) Block diagram
As shown in the figure, the encoder is placed after the
transmitter(TX) circuit and the decoder block is placed at the
end of the channel before the receiver(RX) circuit. The binary
input data is encoded and there are four levels of signals at Q.
In this analysis, 1024 bits have been simulated. 𝑉𝑄 is the
encoded multi-level voltage steps at node Q. [1]
𝑉𝑄 = 𝑉𝑏𝑖𝑡𝑊 (1)
The encoded signals propagate through the channel to node R.
The CHS encoding static matrix, W, spreads each binary across
the line N with specific properties. Thus, the noise that is
coupled from aggressor to victim lines becomes a part of the
41
signals. Then, the data arriving at node R is the encoded
channels at node Q convolved with the impulse response of the
channel. The signals are sampled and recovered at node S by
decoding circuit, 𝑉𝑠
Assuming 𝑉𝑄 = 𝑉𝑅 (with low channel loss and low-reflections),
𝑉𝑆 = 𝑉𝑅𝑊−1 (2)
III. TRADITIONAL ROUTING AND CHS 3D NOVEL ROUTING
The CHS concept has been proved on microstrip routing.
Simulations correlate well with validation [1]. Fig. 2 shows the
eye diagram for a 4-coupled microstrip transmission line having
4-mil spacing between the signal channels at a data rate of 8
Gbps.
Fig. 2 Four-coupled microstrip and eye diagrams with 4 mil trace spacing, six-
inch trace length at 8 Gbps data rate. Binary Traditional Signaling (Red), CHS
(Blue)
Binary traditional signaling requires more than 16 mil spacing
between signals in order to have a good eye opening as
illustrated in Fig. 3. Besides that, CHS is 2.5X faster than binary
signaling and the routing is 2.3X denser [1].
Fig. 3 Comparison between Traditional Binary Signaling (Red) and CHS at
different trace spacing at 8 Gbps
This research has been conducted on stripline routing and CHS
3D Novel routing to investigate the sensitivity of CHS
technique. Both routings schemes show that the CHS concept
is still valid and has advantages compared to traditional binary
signaling. Based on this analysis, CHS can support up to 1 mil
spacing at a conventional dielectric height. By introducing
Novel Coplanar Routing to CHS microstrip routing (~2.3X
compared to traditional binary signaling) and CHS 3D Novel
Routing (~10X compared to traditional binary signaling) the
routing density will double.
However in the CHS concept, conductor one shows eye
degradation for nibble-to-nibble analysis due to common-mode
signaling that is susceptible to the ground return path. In
addition, the first column of the CHS static matrix is associated
with an orthonormal positive vector value [1].
𝐖 = [
1 1 1 1−1 −1 1 11 −1 −1 11 −1 1 −1
] (3)
𝐖−1 = 𝐖𝐓 = [
𝑊11 𝑊21 𝑊31 𝑊41
𝑊12 𝑊22 𝑊32 𝑊42
𝑊13 𝑊23 𝑊33 𝑊43
𝑊14 𝑊24 𝑊34 𝑊44
] (4)
For conductor one, the common-mode noise is additive thus
causing the eye closure. However, for the other conductors most
of the noise is cancelled due to their differential nature. The eye
closure in conductor one is solvable by placing a ground trace
between nibbles or with isolated spacing between nibbles.
Conductor two to conductor four is somewhat “differential in
nature” that will help to improve signal-to-noise ratio by being
“self-referenced” [1]. Thus, eliminating the reference planes
will result in gains to the bandwidth per volume. Note, the
ground planes and power planes that are required for power and
signal integrity have been ignored in this analysis. This
requirement will be included in future research.
IV. NOVEL COPLANAR ROUTING
The new routing scheme that is based on microstrip routing
by eliminating the reference layer and placing the ground
conductor in the signals layer is illustrated in Fig. 4. This new
routing scheme is called “Novel Coplanar Routing”.
Fig. 4 Microstrip routing with reference plane elimination, “Novel Coplanar
Strips”
The electric fields and magnetic fields are shown in the sketches
below for the even-mode and odd-mode configurations between
microstrip and novel coplanar routing. They show different
field distributions between this routing scheme. The traditional
crosstalk is higher on the Novel Coplanar strip compared to the
microstrip routing as shown in Fig. 6 since the fields are
concentrated partially in air. However, for the Novel Coplanar
strip, crosstalk can be reduced by having tight coupling with the
reference signal since the electric field dispersed in the air
42
degrades travelling through different media that close to a
homogenous structure.
Fig. 5 Artist’s concept for electric and magnetic field intensity lines comparison
between Microstrip and Novel Coplanar Strips
Fig. 6 Far end crosstalk comparison between Microstrip, Novel Coplanar Strip
in frequency domain based on two-couple transmission line
The following is an analysis based on Fig. 4 routing. In this
routing scheme, we will expect different impedance values
between traces. Conductor one and four will have lower
impedance values while conductor two and three will have
higher impedances since they are farther from a reference plane.
For a six-inch transmission line, the eye diagram shows
degradation on conductor three compared with a three-inch
channel. This is due to a termination mismatch between signals
that produces more reflection and skew. Based on the CHS 3D
Novel routing approach, this can be solved either by reducing
the channel length or by impedance matching on the routing
scheme.
(a) Three-inch channel length (b) Six-inch channel length Fig. 7 Eye diagram based on Fig. 4 at 8 Gbps data rate (Red- Traditional Binary
Signaling, Blue – CHS)
The eye closing at conductor one is expected due to common-
mode noise that was introduced in the CHS matrix. A reference
signal has been added between signals to solve the reflection,
skew and common-mode issues as illustrated in Fig. 8. This
routing scheme has a symmetrical configuration compared to
Fig. 4 which is an asymmetric configuration. The routing is
close to microstrip routing but offering better density since the
number of layers for the stackup is reduced. The eye diagram
on Fig. 9 shows a better eye opening compared to Fig. 7. In
addition, there is an insignificant impact on the eye opening on
the ten-inch channel length.
Fig. 8 Alternative Novel Coplanar Routing for symmetric configuration
(a) six-inch channel length (b) ten-inch channel length
Fig. 9 Eye diagram at 8 Gbps based on Fig.8 configuration. (Red-Traditional
Binary Signaling, Blue- CHS)
Based on existing research, the common-mode noise will
increase at conductor one in a nibble-to-nibble configuration.
However, with symmetric Novel Coplanar routing the issue is
eliminated based on the edgeside and broadside routing scheme
as illustrated in Fig. 10.
Fig. 10 Novel Coplanar Routing for nibble-to-nibble configurations
This configuration provides a better Return on Technology
Investment (ROTI) compared to traditional routing that will be
shown in the next research paper. Good eye opening has been
observed based on Fig.10 edge-side nibble-to-nibble
configuration as illustrated in Fig. 11 with 2 mil spacing
between signals.
43
Fig. 11 Eye diagram for edgeside nibble-to-nibble configuration at 8 Gbps data rate, ten-inch channel length and 2 mil trace spacing. (Red- Traditional Binary
Signaling, Blue-CHS)
Based on the above analysis, the concept can be extended to
stripline routing as shown in Fig. 12, which demonstrates that
this concept is feasible for this analysis except that the reference
signals need to be added between the common-mode signals.
The stripline configuration shows less crosstalk since the fields
will fringe in the substrate area, which is close to a
homogeneous configuration. In the Novel Coplanar routing
scheme, termination is not an issue and length mismatch can be
supported for up to 100 mils between the signal traces.
Fig. 12 Coplanar Strip approach on stripline nibble-to-nibble routing
In existing research, CHS shows the capability to support the
CHS 3D Novel routing scheme, and can potentially grow an
infinite n number of nibbles vertically or horizontally. There is
the potential that the Novel Coplanar Routing concept can be
applied to 3D CHS Novel routing by eliminating the reference
layer and thus achieving 2X higher bandwidth per volume.
V. CONCLUSION
The CHS concept paves the way for introducing new routing
schemes that will offer higher bandwidth per volume compared
to traditional routing with no design changes required on the
CHS scheme. This paper shows that with this new routing
scheme, the routing density about doubles compared to CHS 3D
Novel routing and CHS microstrip routing. Traditional routing
is not recommended for high speed designs due to high
crosstalk from reflection, and the existing crosstalk solution
does not help reduce the routing density compared to the CHS
method. Nevertheless, this novel routing scheme is relatively
new and further investigation is required to understand the
advantages and sensitivity of the CHS concept. In addition,
proofs of concept need to be conducted in the future. Besides
that, Return on Technology Investment (ROTI) is a new
concept that will be discussed in the next research paper, which
will include the advantages of new routing schemes such as
Novel Coplanar and CHS 3D Novel routing compared to
traditional routing.
ACKNOWLEDGMENT
Special thanks to Professor Dr. Paul G. Huray the author of
the books, Maxwell’s Equations and The Foundations of Signal
Integrity, Dr. Femi Oluwafemi, Stephen H. Hall from Intel,
USA and Tom McDonough from USC for their valuable
guidance throughout this research.
REFERENCES
[1] C. Sreerama, “Novel crosstalk mitigation solutions for high-speed
interconnects to maximize bus band-width and density,” 8th annual signal
integrity symposium, Penn State Harrisburg, PA, Apr. 4, 2014, pp. 1. [2] F. Broyde and E. Clavelier, “A new method for the reduction of crosstalk
and echo in multiconductor interconnections,” IEEE Trans. Circuits Syst.
I, Reg. Papers, Vol. 52, pp. 405-416, Feb. 2005. [3] C. R. Paul, “Analysis of multiconductor transmission lines,” 2nd ed., New
York, NY, Wiley-Interscience, 2007.
[4] S. H. Hall, H. L. Heck, “Advanced signal integrity for high-speed digital designs,” 1st ed., Hoboken, NJ, Wiley-IEEE Press, 2009.
[5] C. Yongjin, H. Braunisch, K. Aygun, and P. D. Franzon, “Analysis of
inter-bundle crosstalk in multimode signaling for high-density interconnects,” ECTC, Lake Buena Vista, FL, 2008, pp. 664-668.
Azniza Abd Aziz received her PhD. In electrical engineering
(signal integrity) from University of South Carolina. Her
current research interest includes Signal Integrity solutions for
high-speed data design. She was an Advanced Signal Integrity
Engineer in Intel, Penang, Malaysia and Senior Signal Integrity
Engineer in Hewlett Packard Enterprise, California, USA with
almost 10 years experiences working on designing, validation
desktop, mobile and server platforms. Currently, she is lecturer
at USM, Malaysia.
Chaitanya Sreerama is a staff hardware engineer at Intel Labs
in Hillsboro, OR. She received her B.S degree in electronics and
communication engineering form JNTU, India in 2001, M.S
degree in electrical engineering (numerical & computational
electromagnetics) from Clemson University in 2004, and PhD.
degree in electrical engineering (signal integrity) from
University of South Carolina in 2014. She has been working at
Intel since 2004, and her areas of expertise include EMI, RFI,
and Signal Integrity. In addition to patents (14) and publications
(4), she has been awarded the 2010 Intel Achievement Award
for her contributions to the company.
44
A Grounded Capacitance Multiplier Based on CCII
J. Vavra Univ. of Defence, Dept. of Electrical Engineering, Kounicova 65, 662 10 Brno, Czech Republic
Abstract—The capacitance multiplier is an active block which
emulates the synthetic passive capacitor, whose capacitance is
several times bigger than the reference real capacitance used in
circuit connection. Multipliers are used in the design of
integrated circuits for the realization of high-value on-chip
capacitances. The paper describes one possible way of
multiplying the capacitance with a minimum number of active
and passive components. The proposed circuit uses only one
active element, the second-generation current conveyor (CCII),
and only one grounded passive component – the reference
capacitor, whose capacitance is being multiplied. This idea
connects the circuit theory and microelectronic approach because
capacitance multipliers are using in the design of integrated
circuit area, where the capacitance multiplier can save valuable
space on the chip. For test reasons, the active element is
composed of commercial amplifiers and the verification is
ensured by experimental measurement on the prototype and by
SPICE simulations.
Index Terms— Capacitance Multiplier, Current Conveyor,
Howland current pump, current mode.
I. INTRODUCTION
The capacitor is one of the basic components of analog
integrated circuits, significantly contributing to the accuracy in
design functionality. The implementation capacitors on the
chip require a large area, especially if the capacitor value must
be as big as possible, like the so-called Zero Making Capacity
in Phase Locked Loop constructions [1]. Moreover, the
fabrication of low-noise capacitors of the same dimensions
(CMOS image sensor) brings further technological
complications not only because of limited space on the chip
[2]. Thus the Capacitance Multipliers (CM) play a very
important role in the area of integrated circuits because they
reduce the requirement for die area because they emulate a
larger capacitance using a much smaller capacitor. In the
0.18μm technology, the MIP capacitors (metal-insolator-
polysilicon) [3] require a typical value of 500 μm2/pF. Taking
into account the high absolute tolerance (±20% in the same
technology), correction mechanisms such as trimming are
often required, leading to an even larger circuit area.
There are two different approaches in the CM design, the
voltage and the current approach. The voltage-mode
capacitance multiplier uses the Miller effect [4-6] through
which a high multiplication factor can be achieved but the
circuit is limited to low-frequency operation and low dynamic
range. The current-mode multipliers add a scaled copy of the
current flowing through a reference capacitor to the driving
circuit, simulating a higher total load current, while
maintaining the load impedance's capacitive nature. This is
equivalent to a larger load capacitance.
Recently many proposals of capacitance multipliers using
active building blocks have been reported. Several of them use
as the active building block the second-generation current
conveyor (CCII) [8-12] or other active elements [13-20].
Many of the proposed ideas fail in one or more of the
following aspects: use of two or more active elements,
excessive employment of passive components and floating
passive components or impossibility of tunability.
The main idea of the paper is a maximum simplification of
CM. The paper offers a simple circuit idea of CM with only
one CCII and one grounded working capacitor. This means
one buffer and one current-controlled current source. The
multiplication can be provided by using an input voltage
amplifier or by amplifying the controlled current source (the
ratio of transistors in the current mirror).
II. SECOND-GENERATION CURRENT CONVEYOR (CCII)
CCII is a well-known and often used active building block
which was introduced by Sedra and Smith in 1970 [21]. Fig. 1
(a) shows its schematic symbol and Fig. 1 (b) its
corresponding behavior model.
Figure 1: Schematic symbol and behavioral model of CCII-.
There are two versions of CCII – inverting and
noninverting, which are labeled CCII+ and CCII- respectively,
depending on the direction of output current Iz. In this paper,
the proposed CM uses the CCII- type.
The voltage signal Vy, which is connected to the input
terminal y, is buffered into the low-impedance input x. The
current flowing into the terminal x as current Ix is conveyed to
the high-impedance output z as the output current Iz flowing
into the terminal z (CCII+) or out from the terminal z (CCII-).
The ideal behavior of the CCII- can be described according to
Fig. 1 by a system of equations (1) as follows:
45
1 0 0
0 0 0
0 1 0
x y
y x
z z
V V
I I
I V
=
. (1)
III. PROPOSED CIRCUIT
The proposed CM circuit based on CCII- is shown in Fig. 2;
it is one of the easiest versions of CMs. The ideal function of
CCII does not allow multiplication, because Vx = Vy and Iz = Ix,
so the multiplication factor is 1 – in the ideal case. But if the
CMs are mainly designed for integrated circuit application, the
construction of current mirror or voltage buffer allows
introducing the multiplication coefficients Ab and/or Ac during
the matching. Referring to this idea, Fig. 2, and using (1) the
corresponding flow graph can be calculated as shown in Fig.
3, where the variables Ab and Ac are the gain of input buffer
from Vy to Vx and the gain of the current mirror from Ix to Iz,
respectively. In the common case, these variables are equal to
1, but this can be changed during the preparation of the final
layout.
Figure 2: Capacitance Multiplier based on CCII-.
Figure 3: Corresponding flow graph of proposed CM.
According to Mason’s rule, the input impedance of the
circuit can be derived as follows:
11
1
1
b c
VZ
I sCA A= = . (2)
The equivalent capacitance Ceq follows from (2) and is
given by the following multiplication:
eq b cC CA A= . (3)
The value of Ceq can be set by the value of voltage gain Ab
or by the value of current gain Ac.
Because of the limited dynamic range, the current multiplier
is more suitable than voltage gain.
IV. ANALYSIS OF REAL EFFECTS
In addition to the ideal parameters, each active element
represents some real features which affect the ideal behavior
of the whole circuit. Some of them can play an important role
in interesting frequency range, dynamic range, or in linearity,
i.e. primarily parasitic impedances of each terminal, finite
voltage tracking error b from the high impedance input
terminal y to the low impedance terminal x and finite current
tracking error c from this terminal x to the high impedance
output terminal z. The frequency dependence of voltage and
current gains is not considered because of their application in
the higher frequency range. The final layout of the chip must
be modified so that these effects will not affect the function of
the application. In Fig. 4, all the parasitic influences
considered are shown in red.
Figure 4: Proposed CM with considered parasitic influences of the CCII-.
The parasitic impedances of the terminals y and z work in
parallel connection, thus they can be connected to one pair of
impedances Ryz = Ry || Rz and Cyz = Cy + Cz.
The modified signal flow graph respecting the real effects is
given in Fig. 5:
Figure 5: Signal flow graph of proposed CM with real influences of the CCII-.
The calculation of the overall impedance is routed to a
complicated model composed of an FDNR (Frequency
Dependent Negative Resistor), an inductor, two capacitors and
two resistors. This calculation can be simplified by separating
the coupled impedances Cyz and Ryz, which are connected
directly to the input terminal. The calculation is then very
simple because it is composed of voltage and current gains Ab,
Ac and their real deviations b c multiplied by a serial
46
connection of the parasitic resistor Rx and the reference
capacitor C. This is suggested by the straight line of the signal
flow graph in Fig. 5.
The input admittance of the circuit in Fig. 4 can be written
as follows:
1
b b c cyz yz
x
sCA AY G sC
sCR
= + +
+.
(4)
Equation (4) implies that the admittance Y is made by a
parallel connection of the coupled impedances Cyz and Ryz and
a serial connection of the resistor Rx and the working reference
capacitor C multiplied by the gain variables Ab, Ac and their
real deviations b c. The final equivalent impedance model of
the proposed CM with the parasitic influences considered is
shown in Fig. 6:
Figure 6: Impedance model of proposed CM based on CCII.
where the values of components Cs and Rs are given by the
equations:
s b b c cC CA A = , xs
b b c c
RR
A A = .
(5)
Considering that all the parasitic influences are in their ideal
values (Rx = 0 Ω, Ryz = ∞ Ω, Cyz = 0 F, b = c = 1), then the
resistance Rs equals 0 and the capacitance Cs equals Ceq in
equation (3).
V. CCII IMPLEMENTATION
As mentioned in the Introduction, the CM is most useful in
the area of integrated circuits, where the realization of real
capacitances is quite expensive because of their size on the
chip. For a better verification of the proposed circuit, the
implementation of CCII is required. It must be clear that the
simulation results presented are limited by the real behavior of
integrated circuits. It can be assumed that the on-chip
realization of CCII will generally have better properties,
depending on the CMOS technology used and on matching.
Figure 7: Conception of the CM based on CCII-.
For verification reasons, an interesting implementation of
CCII [22] can be used. One approach, based on the
well-known Howland current pump, is presented in [23]. The
final proposed circuit of CM with the Howland current pump
as a CCII is shown in Fig. 7.
Because of the negative feedback of OA1, the voltage on the
working capacitor C precisely follows the input voltage V1,
which is the first basic feature of CCII. The current flowing
through the reference capacitor is forced to flow through the
first sensing resistor RS1. The instrumentation amplifier IA1
senses the voltage drop at this resistor, whose output voltage is
given by the sum of this voltage drop and the voltage level at
the ref terminal. The voltage at the ref terminal is a buffered
copy of the input voltage V1, thus the voltage drop at the
resistor RS2 is only given by the voltage drop at the resistor
RS1, which is sensed by IA1. The result is the input current I1,
which is formed by the current flowing through the reference
capacitor multiplied by the ratio of the sensing resistors RS1
and RS2 assuming the unity gain of IA1 – if RG is omitted:
1
2
Sc
S
RA
R= .
(6)
In Section III, the possibility of tuning the gain
(multiplication factor of CM) is presented. In the
implementation presented, two possible ways of changing the
gain are available. The voltage gain Ab can be changed by
adjusting a higher gain of the instrumentation amplifier IA1.
Thus the RG must be added, otherwise the unit gain of the
instrumentation operating amplifier IA1 is preset. Another
possible way to increase the gain is by changing the ratio of
resistors RS1 and RS2. It can be considered as a current gain Ac.
VI. PRINCIPLE VERIFICATION
For a concrete example, the working capacitance can be
selected C = 1 nF and the multiplication coefficient can be 10
(Ab = 1, Ac = 10). According to (6), the resistor RS1 = 10RS2.
With respect to the limited dynamic range, which is affected
by the value of RS1, the values of these resistors are selected as
follows: RS1 = 33 kΩ and RS2 = 3.3 kΩ. All the amplifiers in
the proposed circuit were selected with respect to giving
preference to accuracy, low power consumption, and low
offset. The precision instrumentation amplifier AD8226 [24]
was selected as IA1. The zero drift CMOS operational
47
amplifiers OPA735 [25] were selected as both operating
amplifiers OA1 and OA2. All active devices were supplied by
±5 V. The amplifiers AD8226 and OPA735 provide low offset
voltage (50 V and 5 V) with low drift (0.5 μV/°C and 0.05
μV/°C). The input differential impedance of the AD8226 is
800 M with a parasitic capacitance of 2 pF, thus it does not
degrade the output resistance of the CCII. The low input bias
current (200 pA) of the OPA735-based voltage buffers
represents a negligible error in setting the output current.
The parameters of the final equivalent circuit in Fig. 6 can
be obtained from a detailed analysis of the Howland current
pump, as is done in [26]. According to this analysis, these
parameters are as follows: Cyz = 4 pF, Ryz = 800 kΩ,
Rs = 0.01 Ω, Cs = 9.99 nF according to (5) and for
low-frequency signals.
For a verification of the proposed circuit, the RC circuit in
series connection was selected. The CM was connected to the
input signal VIN = 1V/10kHz through the resistor R = 1 kΩ.
The operation of CM was tested via transient analysis and
measured on a sample which was assembled using
components as mentioned above. The results of measurement
and the SPICE transient analysis of the proposed RC circuit
for 1 V/10 kHz sinusoidal excitation are shown in Fig. 8. The
ideal waveform is the same as a simulated response. As is
clear from Fig. 8, the amplitude of the measured current
response flowing through the real CM is practically the same
as the current response flowing through an ideal capacitor with
a capacitance of 10 nF and simulated CM. The small visible
deviations are caused by the parasitic resistor Rs.
Figure 8: Measured waveform (I1M) of proposed RC circuit with CM for
1 V/10 kHz sinusoidal excitation in comparison with simulated (I1S) and ideal (I1I) current response.
VII. CONCLUSION
The paper describes the utilization of capacitance
multipliers in a microelectronic area. The capacity of
grounded reference capacitance is transformed to a higher
value by using only one basic active element and one
grounded capacitor. This transformation saves expensive
space on the chip. Disadvantages of this transformation are the
implementation of parasitic impedances and frequency
limitations which are hidden in the active element. The
analysis points out the possibility of eliminating some
parasitic influences, especially in terms of mere changes of the
cutoff frequency.
The proposed capacitance multiplier was analyzed for real
effects, measured on the real specimen and simulated via the
SPICE transient analysis. The measured and simulation results
are in good agreement with the behavior of ideal CM.
ACKNOWLEDGMENT
This work was supported by the Project for the development
of K217 Department, University of Defence Brno.
REFERENCES
[1] J. Choi, J. Park, W. Kim, K. Lim and J. Laskar, “High multiplication
factor capacitor multiplier for an on-chip PLL loop filter,” Electronics
Letters, vol. 45, no. 5, pp. 239–240, 2009. [2] E. Zadeh, “CM circuits and the applications thereof to attenuate row-
wise temporal noise in image sensors,” United States Patent,
20080246539 A1, Oct 9, 2008. [3] F. Maloberti, Analog Design for CMOS VLSI Systems, 1st Ed., New
York, United States of America: Springer, 2001. [4] P.R. Gray et.al., Analysis and Design of Analog Integrated Circuits, 4th
Ed., New York, United States of America: John Wiley & Sons, 2001.
[5] Gabriel A. Rincon-Mora, “Active Capacitor Multiplier in MillerCompensated Circuits,” IEEE Transactions On Solid-State
Circuits, Vol. 35, No. 1, Jan. 2000.
[6] B. Razavi, Design of Analog CMOS Integrated Circuits. Tata McGraw Hill Edition 2002.
[7] Y. Tang, M. Ismail, and S. Bibyk, “Adaptive Miller capacitor multiplier
for compact on-chip PLL filter,” Electronics Letters, vol. 39, no. 1, pp. 43–45, 2003.
[8] M.T. Abuelma’atti, and N.A. Tasadduq, “Electronically tunable
capacitance multiplier and frequency-dependent negative-resistance simulator using the current-controlled current conveyor,”
Microelectronics Journal, vol. 30, pp. 869–873, 1999.
[9] A.A. Khan, S. Bimal, K.K. Dey and S.S. Roy, “Current conveyor based R- and C-multiplier circuits,” AEU-Int. J. of Electronics and
Communications, vol. 56, no. 5, pp. 312–316, 2002.
[10] A. Lahiri, “DO-CCII based generalized impedance convertor simulates floating inductance, capacitance multiplier and FDNR,” Australian. J. of
Electrical and Electronics Engineering, vol. 7, no. 1, pp. 15–19, 2010.
[11] P. Prommee, and M. Somdunyakanok, “CMOS-based current-controlled DDCC and its applications to capacitance multiplier and universal
filter,” AEU-Int. J. of Electronics and Communications, vol. 65, no. 1,
pp. 1–8, 2011. [12] I. Myderrizi, and A. Zeki, “Electronically tunable DXCCII-based
grounded capacitance multiplier,” AEU-Int. J. of Electronics and
Communications, vol. 68, no. 9, pp. 899–906, 2014. [13] C.K. Alexander and M.N.O. Sadiku, “Fundamentals of Electric
Circuits,” McGraw Hill, 5th edition, 2013, pp. 437–439.
[14] W. Jaikla and M. Siripruchyanan, “An electronically controllable capacitance multiplier with temperature compensation,” In Proc. Int.
Symp. on Communications and Information Technologies ISCIT 2006,
2006, pp. 356–359. [15] A. Jantakun, N. Pisutthipong and M. Siripruchyanun, “Single element
based novel temperature insensitive/electronically controllable floating
capacitance multiplier and its application,” In Proc. Int. Conf. on Electrical Engineering, Electronics, Computer, Telecommunications and
Information Technology ECTI-CON 2010, 2010, pp. 37–41.
[16] P. Silapan, C. Tanaphatsiri and M. Siripruchyanan, “Current controlled CCTA based-novel grounded capacitance multiplier with temperature
compensation,” In Proc. of Asia Pacific Conf. on Circuits and Systems
APCCAS 2008, 2008, pp. 1490–1493. [17] G. Ferri, and S. Pennisi, “A 1.5-V current-mode capacitance multiplier,”
In Proc. 10th Int. Conf. on Microelectronics ICM’98, 1998, pp. 9–12.
[18] M. Siripruchyanan, and W. Jaikla, “Floating capacitance multiplier using DVCC andCCCIIs,” In Proc. Int. Symp. on Communications and
Information Technologies ISCIT 2007, 2007, pp. 218–221.
48
[19] H.Y. Darweesh, F.A. Farag and Y.A. Khalaf, “New active capacitance
multiplier for low cutoff frequency filter design,” In Proc 19th Int. Conf.
on Microelectronics ICM’07, 2007, pp. 381–384.
[20] J. Vavra, “A capacitance multiplier based on DBTA,” In Proc 2017 IEEE Nordic Circuits and Systems Conference (NORCAS): NORCHIP
and International Symposium of System-on-Chip (SoC), Linkoping,
2017, pp. 1–5. [21] A. Sedra and K. Smith, "A second-generation current conveyor and its
applications," IEEE Transactions on Circuit Theory, vol. 17, no. 1, pp.
132–134, Feb. 1970. [22] E. Tlelo-Cuautle, Integrated Circuits for Analog Signal Processing. New
York: Springer, 2013.
[23] TEXAS INSTRUMENTS. A Comprehensive Study of the Howland Current Pump. AN-1515. Texas Instruments, Post Office Box 655303,
Dallas, Texas, 2013.
[24] AD8226, Wide Supply Range, Rail-to-Rail Output Instrumentation Amplifier. Datasheet, Analog Devices, Rev. C, 2017.
[25] OPA735, 0.05V/°C max, SINGLE-SUPPLY CMOS OPERATIONAL
AMPLIFIERS Zero-Drift Series. Datasheet, Texas Instruments, Rev. Feb. 2005.
[26] D. Biolek, J. Vavra, “Precise Implementation of CDTA,“ Advances in
Electrical and Electronic Engineering, vol. 15, no. 5, pp. 824–832, Dec. 2017.
49
An Overview of Information Hiding
Techniques; Challenges and Applications
A. Omar Adil Deheyab1, 2, B. B Rahmatullah22 and C. M. Hashim33
1Universiti Pendidikan Sultan Idris , Malysia. Corresponding author: First A. Author (e-mail: [email protected]).
Abstract ; Currently, many people communicate with
each other over the Internet, however, information
transmission is not very secure, where there is a high
probability for copying and altering the information
easily. Transmitting the sensitive or private information
needs some kind of protection especially in the public
network unless a secure channel is utilized for the
transmission. Information hiding is one of efficient
technique that provides solution to the problem of
transmitting important data over communication
channels. These techniques should meet various
applications requirements such as high robustness
against attacks, real-time, and high hiding capacity.
Generally, information hiding techniques suffer from
many challenges. This paper seeks to provide an insight
of these techniques, their applications and challenges to
fill the gap of missing such overview.
Keywords: Information hiding; Steganography;
Cryptography; Watermarking
1. Introduction
The Internet is an open network so it is easy to obtain
and transmit all kinds of multimedia information like
audio, video and images freely, which brings many
threats to information security such as piracy, theft,
and ownership issues [47] . Several methods have
been utilized to protect sensitive and critical
information during their storage or transmission.
They mainly depend on cryptography or information
hiding techniques (also called data hiding) [3].
Information hiding is the art of embedding
information inside another medium during
transmission. The information is embedded into a
cover medium (image, audio, video, or text) to create
the embedded medium (in watermarking applications
is called watermarked object, while in steganographic
applications is called a stego object) [14]. The
information hiding scheme requires visual quality of
embedded images, hiding capacity (called payload),
and robustness. The scheme with low image
distortion is more secure than that of high distortion
because it does not raise the attackers’ suspicions.
The scheme with a high payload is preferred because
it assists to transmit more secret data. The robustness
is mainly significant but achieving robustness is
technically challenging in high-payload data hiding
scheme. Generally, visual quality, hiding capacity,
and robustness are conflicting issues; therefore, a
tradeoff among them is desirable. However, the
tradeoff differs from application to application,
depending on users’ requirements and application
domains [14]. Information hiding techniques can be
divided into irreversible and reversible information
hiding. If a data-embedding scheme is irreversible
(also called lossy), then the secret data only can be
extracted and the original cover image cannot be
restored. On the other hand, a reversible (also called
invertible, lossless, or distortion-free) data-
embedding scheme allows recovering the original
cover image completely upon secret data extraction.
This technique plays a vital role in many
applications, such as military and medical images
[14,46, 34] . Information hiding techniques (called
information hiding or information embedding) can be
categorized into two types: steganography and
watermarking according to the applications that
information hiding is utilized for [48, 3].
Cryptography is considered as a hiding technique. It
can be interwoven with steganography and
watermarking. The goal of cryptography and
steganography is to conceal the data, but their
implementation methodologies are varied. The
watermarking and steganography methodology is
same, but their goals are varied. Watermarking
concerned with digital data copyright protection,
while steganography deals with digital information
hiding[5].
2. Steganography
Steganography is a term derived from the Greek
word steganos which means covered or secret and
graphie that means writing or drawing i.e., covered
writing [2,10]. Steganography refers to hide
50
information in a cover medium without changing its
original quality to prevent its detection from
unauthorized attempts which makes it a main choice
for secret communication [12,45,2]. In other way,
steganography prevents the intruder from suspecting
the secret information in the cover object. The cover
medium (also called host or carrier) usually may be
any digital medium such as an image, audio, or video
file. However, images are widely used as a cover for
steganography because of its prevalence in daily
applications and high redundancy in representation
[27,35]. The hidden message (also called payload)
may be of any type such as text, image, audio, or
video [21,17]. The two important properties of
steganography are (i) good visual/statistical
imperceptibility of the payload which is essential for
security of hidden communication and (ii) payload is
essential to convey large quantity of secret
information. Steganography has to satisfy capability
and the transparency requirements. Capability means
embedding large payload into media. Transparency
indicates an ability to prevent distinctions between
stego and cover image by perceptual or statistical
analysis [38]. The security of a steganography system
increases if the payload remains unreadable to an
attacker even if he has knowledge about the
embedding method [35]. Steganography techniques
can be classified into two major categories such as
spatial domain techniques and transform (frequency)
domain techniques. In spatial domain techniques the
secret message is hidden inside the image by
applying some manipulation over the image various
pixels [45]. In transform domain techniques the
image is transformed to frequency domain and the
secret message is hidden in the coefficients [1].
Steganography methods in spatial domain include
LSB steganography, RGB based steganography, pixel
value differencing steganography, mapping based
steganography, palette based steganography, collage
based steganography, spread spectrum
steganography, code based steganography, and others
[45]. Steganography methods in transform domain
include DCT (Discrete Cosine Transfer), DWT
(Discrete Wavelet Transfer), and DFT (Discrete
Fourier Transfer) [40]. Spatial domain techniques
offer higher payload but are prone to various normal
attacks such as JPEG compression, noise attacks, and
low-pass/high-pass filtering and geometric attacks
such as image resizing, cropping and rotations by
different angles. Transform domain techniques
provide lower payload but can resist various attacks
[32].
2.1 Steganography Challenges
The major challenge in steganography is how to
produce stego images with high imperceptible, and
how to increase the amount of payload capacities in
the stego image [4]. The message must be hidden in
the cover image in such a way that the generated
stego-image does not deviate much from the original
image, visually and statistically [17]. Steganalysis is
devoted to defeat steganography and detect its
presence. Therefore, steganography must pay more
attention to the visual quality, the statistical
imperceptibility, the capacity of embedded data, and
the resistance against detection [48,31]. Visual
quality is a significant issue, in some applications
such as law enforcement, military image systems, and
medical diagnosis, where a small image distortion is
unacceptable [14].
Another significant challenge is the limited
embedding rate in the transform domain, and the
vulnerability of spatial domain to various attacks like
JPEG compression, high pass filtering, low pass
filtering, cropping etc [21,16,4]. There are many
steganography algorithms proposed by many
researchers, however, some of these algorithms are
very complicated due to the long time needed to hide
secret data such as DWT, while the others are simple
methods with low complexity as in LSB [4]. It is
worth mentioning that steganalysis tools development
degrades steganography schemes performance, thus
researchers should develop secure steganography
techniques to defeat attackers and steganalysis [31].
2.2 Steganography Applications
Steganography has different convenient applications.
The applications include copyright control of
materials, enhancing robustness of image search
engines and smart IDs (identity cards). Other
applications are video-audio synchronization,
companies’ safe circulation of secret data, TV
broadcasting, TCP/IP packets, medical imaging
[15].The steganography can also be used for secure
exchange of secret messages between sensitive
organizations, securing online banking, and voting
systems, nefarious use by attackers to send viruses
and Trojan horses, as well as secret communication
between terrorists and criminals [32].
3. Watermarking
The development of the computer and network
technology facilitates the production, distribution,
acquiring and copying digital media products such as
digital text, image, video and audio. Therefore, it is
mandatory to protect digital media copyright of the
media owner and consumer [28]. One of the effective
solutions is the digital signature, which can be
51
embedded in data by using watermarking. Digital
watermarking is an effective solution for copyrights
protection with the extensive productions of digital
multimedia in the information technology era [29].
Digital watermarking is one of the common methods
for authentication digital images. Digital
watermarking inserts a sign or logo into the cover
image that can confirm the originality of the image
after extraction [8]. Digital watermarking concept is
similar to steganography; they both hide a data inside
a digital media. However, the difference between
them is their goal. Watermarking hides a data related
to the actual content of the digital signal where both
data and digital signal are important, while
steganography has no relation between digital signal
and data. The digital signal used as a cover to hide
the data i.e. the data is mainly important only [26].
Hiding the watermark is not necessary for all
scenarios, and high capacity is also not necessary as
signatures are usually small. Generally, watermarking
protects digital data from copyright violation and
counterfeiting, by embedding a digital signature with
less distortion in the cover image. However, this
distortion is not acceptable in case of military and
medical images [14,47]. Moreover, watermarking
schemes are also desirable for protection of the
content and the integrity of images due to their ability
of recovery, as well as tamper detection [7]. The
watermarking can be visible or invisible. In visible
watermarking technique, human can perceive the
watermark, while in invisible watermarking
technique the watermark cannot be perceived by
human. The invisible watermark is established by
alteration made to the pixels that cannot be
perceptually noticed [19]. There are three basic
requirements for invisible watermarking. One of
them is that the distortion to the pixels of host image
(due to embedding of watermark) should be too small
to be noticed. The second requirement it should be
robust in case of various attacks, and finally high
security should be provided to the watermarking
scheme [30].
Watermarking technologies are categorized
according to implementation domain to spatial
domain or frequency (transfer) domain. In spatial
domain, data embedding is performed directly
through pixel modification, while in frequency
domain, data is embedded in the coefficient. The
transfer domain schemes are more computational
complexity comparing to the spatial domain but
provide better robustness than spatial domain
schemes with lesser payloads [43,9]. As the
watermarking is considered as an information hiding
technique, it can also be divided into irreversible and
reversible. The reversible watermarking scheme must
have the ability to recover the watermark even if the
watermarked image has undergone un-intentional
attacks [46]. Digital image watermarking generally
classified in three main groups Robust, Fragile and
semi-Fragile [33,7,19,8]. Robust watermark has the
ability to resist certain malicious attacks or general
image processing operations like cropping, filtering,
compression, etc. [50,51]. It is used for copyright
protection and finger printing applications. Fragile
watermarking is very sensitive to any changes to the
watermark image. It will be destroyed if its content is
slightly tampered. It is used for authentication and
tamper detection. Semi fragile watermarking has the
ability to endure some un-intentional changes such as
the lossy compression and channel noise but it is
fragile against malicious attack [16,42,40].
3.1 Watermarking Challenges
The robustness of digital watermarking schemes is
critical [14]. It is also critical to develop general
technique for all the digital media that is robust
against various attacks [16]. Generally, achieving the
watermark requirements in terms of imperceptibility,
robustness, and high capacity is necessary. However,
it is difficult to maintain balance between
imperceptibility, robustness and capacity [16]. Some
other challenges of watermark are identified as
follows [2]:
• It is destroyed if anyone manipulates the
image.
• The watermark may become unreadable due
to some operations like compressing,
resizing
• It protects the copy right but doesn’t prevent
copying of image. However, it can trace and
detect the copied image ownership.
Although watermarking is an easy tool to protect
ownership of data, but a lot of fraudulent cases are
reported [23].
3.2 Watermarking Applications
The watermark used as a significant tool in various
applications. Traditionally, watermarking has been
used to certify a document's authenticity for
passports, money and certificates, copyright
protection, document authentication, monitor
broadcast news stories, advertisements and internet
promotions [49]. In the following are different
applications for digital watermarking [43,25,31]
• Copyright Protection: to identify the image
source and its authorized user
52
• Data Authentication: Digital signature is a
common data authentication approach
• Fingerprint: increase the stiffness of the
image against alteration or elimination
• Copy Control: keep the security and control
during information distributing and
publishing
• Device Control: used to control access to the
resource using a verifying device
• Fingerprinting: carry the information about
legal recipients
• Tracking: track the users, identify illegal
content, track transactions
• Temper detection: reveal all the changes
which have been made. It is similar to
authentication
• Broadcast monitoring: broadcast the
conversation over some radio and television
• Completeness: detect the modification in
data. This is also an authentication
• Intellectual property protection
4. Cryptography
The term cryptography has come from the Greek
word “kryptós”, which means hidden and “gràphin”
which means “writing”. Therefore, the cryptography
meaning is “hidden writing [2,37]. Cryptography is
the protecting data process by changing it (plaintext)
into an unreadable format called cipher text. The
changing process of data to another format is called
encryption. Decryption is the process of recovering
the original data from the cipher text by using a
secret key. Cryptography is crucial to secure data
from illegitimate access by attaining their
confidentiality, authorization, non-repudiation,
integrity, and availability [17,44]. In cryptography,
the sensitive information is written in such a way that
only the intended recipient can recover it [20]. An
unauthorized party knows the existence of private
data, but his challenge is how to decipher the
encrypted data [17,20]. Cryptography scrambles and
disorganizes information in such a way it cannot be
recognized by attackers during storing and
transmitting. Modern cryptography is heavily based
on mathematical and computer science practice
[38,24]. The cryptography can be divided into
symmetric-key (Private key) cryptography and
asymmetric-key (Public key) cryptography. In
symmetric key cryptography, same key is used by the
sender and the recipient for encryption and
decryption, while in asymmetric key cryptography,
each sender and recipient use two keys, public key
for encryption and a private key for decryption. The
symmetric key cryptography is generally preferred
for large data such as image and video. Asymmetric
key cryptography has higher computation costs.
Examples of symmetric key cryptography are data
encryption standard (DES), triple DES, RC2, RC4
IDEA, AES, Blowfish and Skipjack. Examples of
asymmetric key algorithms are Diffie-Hellman, RSA
and Merkle-Hellman [6,39,40]. Symmetric
algorithms are much faster than asymmetric ones
[18]. The quality of encryption is tested by its
capability to defend different attacks like known
plaintext attack, cipher text only attack, statistical
attack, deferential attack, and brute-force attack, etc
[6]. Cryptography strength depends on the key size,
the more key size; the more expensive computing
power is required to decrypt cipher text [37].
4.1 Cryptography Challenges
Despite outstanding functionalities and security
pledge of cryptography techniques, utilizing
cryptography to provide effective security in practice
is a challenge. Practical settings regularly raise
threats that are not sufficiently reflected by traditional
security modeling of cryptographic [36]. The most
secure cryptographic systems can be rendered fully
insecure by a single specification or programming
error [11]. Although cryptography algorithms are
very useful, it is also breakable. Encrypted data can
sometimes be broken by cryptanalysis, also called
code breaking [44]. All cryptographic based
algorithms have some advantages and disadvantages
related to time and security problems [39]. For
example, traditional symmetric algorithms, such as
DES, AES, and Blowfish require more time, storage
and mathematical computation [13]. Cryptographic
algorithms are cost and computation power intensive.
For example, asymmetric ciphers are computationally
needs more hardware and software [18]. Generally,
cryptographic methods Limitations include 1) it takes
a long time to detect the code 2) huge computational
complexity, 3) cipher form attracts the attackers’
attention, resulting in modification or decryption of
secret data [10,31]. However, cryptography can
prevent the unauthorized access to the important data,
but it cannot prevent the legal person from copying
the decrypted data [25] . Last but certainly not least
while cryptography is very powerful for securing
data; the cryptanalysts could success to break the
ciphers by analyzing the contents of cipher text to get
back the plaintext.
4.2 Cryptography Applications
Cryptography is very significant to provide security
against statistical attacks and other types of attacks
53
when data are exchanged between two parties on a
network [44]. Therefore, it is used for security
purposes in wide applications including [22]:
• Governments’ applications such as
diplomatic missives, military purposes, army
plans.
• Industrial applications such as mobile,
Internet, wireless communication, e-
commerce applications
• Data applications, such as electronic mail,
electronic data interchange (EDI), transfer of
Domain Name System (DNS) and routing
information, electronic forms, and digitally
signed documents, information security
• Financial applications such as payments,
electronic checks, and online banking
5. Conclusion
To protect sensitive and confidential data from illegal
access and illegitimate use during their storage or
transmission through public network like the Internet,
several protection techniques are used such as
steganography, watermarking, and cryptography.
Even these techniques are important and necessary
for various and wide applications, they suffer from
some challenges that affect their performance. This
paper introduces an overview on these techniques and
reviews some of their significant challenges. The
paper also presents different applications for of these
techniques, which are significant to various research
works.
References
[1] Advith, J., Varun, K. R., & Manikantan, K. (2016). Novel
digital image watermarking using DWT-DFT-SVD in YCbCr
color space. In Emerging Trends in Engineering, Technology and
Science (ICETETS), International Conference on (pp. 1-6). IEEE [2] Agrawal, T. (2015). A Survey on Information Hiding
Technique Digital Watermarking. International Journal of
Electrical, Electronics and DataCommunication,Vol.3(8). [3]Al-Dmour, H., & Al-Ani, A. (2016). Quality optimized medical
image information hiding algorithm that employs edge detection
and data coding. Computer methods and programs in biomedicine, 127, 24-43.
[4] Al-Shatanawi, O. M., & El Emam, N. N. (2015). A New Image
Steganography Algorithm Based on Mlsb Method with Random Pixels Selection. International Journal of Network Security & Its
Applications, 7(2), 37.
[5] Anees, A., Siddiqui, A. M., Ahmed, J., & Hussain, I. (2014). A technique for digital steganography using chaotic maps. Nonlinear
Dynamics, 75(4), 807-816.
[6] Ansari, S., Gupta, N., & Agrawal, S. (2012). A Review on Chaotic Map Based Cryptography. International Journal of
Scientific Engineering and Technology
[7] Zhang, J., Zhang, Q., & Lv, H. (2013). A novel image tamper localization and recovery algorithm based on watermarking
technology. Optik-International Journal for Light and Electron
Optics, 124(23), 6367-6371.[8] Barani, M. J., Valandar, M. Y., & Ayubi, P. (2015). A secure watermark embedding approach based
on chaotic map for image tamper detection. In Information and Knowledge Technology (IKT), 2015 7th Conference on (pp. 1-5).
IEEE.
[9] Basu, A., Roy, S. S., & Chattopadhyay, A. (2016). Implementation of a spatial domain salient region based digital
image watermarking scheme. In Research in Computational
Intelligence and Communication Networks (ICRCICN), 2016 Second International Conference on (pp. 269-270). IEEE.
[10] Bhaskar, K., Bakale, M., Chaure, P., & Shirke, P.
(2015).Image Steganography for data hiding Using Huffman code, Zigzag and OPAP..IJETTCS, Vol.4 (6).
[11] Boneh, D., Shoup, V. (2008). A graduate course in applied
cryptography. Version 0.1, from http://cryptobook. net. [12] Bremnavas, I., Poorna, B., & Kanagachidambaresan, G. R.
(2011). Medical image security using LSB and chaotic logistic
map. [13] Brindha, K., & Jeyanthi, N. (2015). Secured Document Sharing Using Visual Cryptography in Cloud Data
Storage. Cybernetics and Information Technologies, 15(4), 111-
123. [14] Chang, C. C., & Kieu, T. D. (2010). A reversible data hiding
scheme using complementary embedding strategy. Information
Sciences, 180(16), 3045-3058. [15] Cheddad, A., Condell, J., Curran, K., & Mc Kevitt, P. (2010).
Digital image steganography: Survey and analysis of current
methods. Signal processing, 90(3), 727-752. [16] Garg, K. (2015). Digital Watermarking: Potential Challenges
and Issues. International Journal of Science, Engineering and Computer Technology, 5(3), 48.
[17] Ghebleh, M., & Kanso, A. (2014). A robust chaotic algorithm
for digital image steganography. Communications in Nonlinear
Science and Numerical Simulation, 19(6), 1898-1907.
[18] Gupta, M. (2015). Security Challenges and Cryptography in
Embedded Systems. IJARSE, Vol.4 (1). [19] Hamouda, K., Elmogy, M., & El-Desouky, B. S. (2014). A
fragile watermarking authentication schema based on Chaotic
maps and fuzzy cmeans clustering technique. In Computer Engineering & Systems (ICCES), 2014 9th International
Conference on (pp. 245-252). IEEE.
[20] Kanso, A., & Own, H. S. (2012). Steganographic algorithm based on a chaotic map. Communications in Nonlinear Science and
Numerical Simulation, 17(8), 3287-3302.
[21] Kaur, D. P., Kaur, J., Deep, K. (2009) Digital Image Watermarking: challenges and approaches for robust algorithm.
International Journal of Electronics Engineering, vol. 1(1), pp. 95-
97. [22] Kessler, G. C. (2016). An Overview of Cryptography
(Updated Version, 3 March 2016). Aeronautical University
[23] Khanna, A. K., Roy, N. R., & Verma, B. (2016, April). Digital image watermarking and its optimization using genetic algorithm.
In Computing, Communication and Automation (ICCCA), 2016
International Conference on (pp. 1140-1144). IEEE. [24] Kumar, N., Deepika, Wadhwa, D., Tomer, D.,
Vijayalakshmi, S. (2014). Review on Different Chaotic Based
Image Encryption Techniques. Intern. Journal of Info. and Computation Tech., Vol. 4 (2), pp. 197-206.
[25] Kumar, S., & Dutta, A. (2016). Performance analysis of
spatial domain digital watermarking techniques. In Information Communication and Embedded Systems (ICICES), 2016
International Conference on (pp. 1-4). IEEE.
[26] Kurdi, M. M., Elzein, I. A., & Zeki, A. M. (2016). Least Significant Bit (LSB) and Random Right Circular Shift (RRCF) in
digital watermarking. In Computer Engineering Conference
(ICENCO), 2016 12th International (pp. 111-116). IEEE. [27] Li, B., He, J., Huang, J., & Shi, Y. Q. (2011). A survey on
image steganography and steganalysis. Journal of Information
Hiding and Multimedia Signal Processing, 2(2), 142-172. [28] Li, S. Q., & Wu, Y. (2010). A Robust Chaos-Based
Watermarking for Copyright Protection. In Biomedical
Engineering and Computer Science (ICBECS), 2010 International Conference on (pp. 1-3). IEEE
54
[29] Mohammadi, S. (2015). A chaotic watermarking scheme using discrete cosine transform. In Information Security and
Cryptology (ISCISC), 2015 12th International Iranian Society of
Cryptology Conference on (pp. 6-10). IEEE. [30] Moniruzzaman, M., Hawlader, M. A. K., & Hossain, M. F.
(2014, May). An image fragile watermarking scheme based on
chaotic system for image tamper detection. In Informatics, Electronics & Vision (ICIEV), 2014 International Conference on
(pp. 1-6). IEEE. [31] Mstafa, R. J.,
Elleithy, K. M., & Abdelfattah, E. (2017). Video steganography techniques: Taxonomy, challenges, and future directions. In
Systems, Applications and Technology Conference (LISAT), 2017
IEEE Long Island (pp. 1-6). IEEE. [32] Muhammad, K., Ahmad, J., Sajjad, M., & Zubair, M. (2015).
Secure image steganography using cryptography and image
transposition. arXiv preprint arXiv:1510.04413. [33] Naseem, M. T., Qureshi, I. M., & Muzaffar, M. Z. (2013,
December). Chaos based invertible authentication of medical
images. In Emerging Technologies (ICET), 2013 IEEE 9th International Conference on (pp. 1-5). IEEE.
[34] Nguyen, T. S., Chang, C. C., & Chang, W. C. (2016). High
capacity reversible data hiding scheme for encrypted images. Signal Processing: Image Communication, 44, 84-91.
[35] Roy, R., Changder, S., Sarkar, A., & Debnath, N. C. (2013).
Evaluating image steganography techniques: Future research challenges. In Computing, Management and Telecommunications
(ComManTel), International Conference on (pp. 309-314). IEEE. [36] Russell, A., Tang, Q., Yung, M., & Zhou, H. S. (2016).
Destroying Steganography via Amalgamation: Kleptographically
CPA Secure Public Key Encryption. IACR Cryptology ePrint
Archive, 2016, 530.
[37] Saleh, M. E., Aly, A. A., & Omara, F. A. (2016). Data
Security Using Cryptography and Steganography Techniques. IJACSA) International Journal of Advanced Computer Science and
Applications, 7(6)
[38] Sathisha, N., Madhusudan, G. N., Bharathesh, S., Babu, K. S., Raja, K. B., & Venugopal, K. R. (2010). Chaos based spatial
domain steganography using MSB. In Industrial and Information
Systems (ICIIS), 2010 International Conference on (pp. 177-182). IEEE. [39] Scholar, P., & Nair, J. S. (2015). A
Review of Image based Cryptography. International Journal of
Computer Security & Source Code Analysis, 1(3), 13-16. [40] Sethi, P., & Kapoor, V. (2016). A Secured System for
Information Hiding in Image Steganography using Genetic
Algorithm and Cryptography. International Journal of Computer Applications, 144(9).
[41] Sheth, R. K., & Nath, V. V. (2016). Secured digital image
watermarking with discrete cosine transform and discrete wavelet transform method. In Advances in Computing, Communication, &
Automation (ICACCA)(Spring), International Conference on (pp.
1-5). IEEE. [42] Shukla, M. K. M., & Mehta, M. A. K. (2015). A Review on Digital Watermarking
Techniques, Applications and Attacks. International Journal Of
Engineering And Computer Science, Vol.4(4). [43] Singh, R. K., Shaw, D. K., & Alam, M. J. (2015).
Experimental Studies of LSB Watermarking with Different Noise.
Procedia Computer Science, 54, 612-620. [44] Singh, S., & Mandoria, H. L. (2017). A Review on Image
Encryption Technique and to Extract Feature from Image.
International Journal of Computer Applications, 163(1). [45] Swain, G., & Lenka, S. K. (2014). Classification of image
steganography techniques in spatial domain: a study. Int. J.
Comput. Sci. Eng. Tech.(IJCSET), 5(3), 219-232.
[46] Thabit, R., & Khoo, B. E. (2015). A new robust lossless data hiding scheme and its application to color medical images. Digital
Signal Processing, 38, 77-94.
[47] Wang, K., Lu, Z. M., & Hu, Y. J. (2013). A high capacity lossless data hiding scheme for JPEG images. Journal of systems
and software, 86(7), 1965-1975.
[48] Wang, P., Wei, Z., & Cao, C. (2010). A Fragile Watermarking Algorithm Based on Logistic System. In Information and
Computing (ICIC), 2010 Third International Conference on (Vol.
4, pp. 304-307). IEEE. [49] Al-Maweri, N. A. S., Ali, R., Wan Adnan, W. A., Ramli, A.
R., Abdul Rahman, S. M. S. A. (2016). State-of-the-Art in
Techniques of Text Digital Watermarking: Challenges and Limitations. Journal of Computer Sciences, Vol. 12 (2).
[50] Nagar, G., & Agarwal, S. K. (2015). Robust digital image
watermarking based on discrete wavelet transform and harmony search algorithm optimization technique. In Applied and
Theoretical Computing and Communication Technology
(iCATccT), 2015 International Conference on (pp. 665-668). IEEE.
[51] Mitashe, M. R., Habib, A. R. B., Razzaque, A., Tanima, I. A.,
& Uddin, J. (2017). An adaptive digital image watermarking scheme with PSO, DWT and XFCM. In Imaging, Vision & Pattern
Recognition (icIVPR), 2017 IEEE International Conference on
(pp. 1-5). IEEE.
55
Virtual Rehabilitation System for Carpal Tunnel
Syndrome Through Spherical Robots
Jorge S. Sánchez, Jessica S. Ortiz, Paola M. Velasco, Washington X. Quevedo, Daniel Castillo-Carrión,
Aldrin Acosta F, Julio Tapia, Cesar A. Naranjo, Franklin M. Silva M and Víctor H. Andaluz Universidad de las Fuerzas Armadas ESPE, Sangolquí-Ecuador
jssanchez, jsortiz4, pmvelasco1, wxquevedo, dacastillo, gaacosta, jctapia3,canaranjo, fmsilva, [email protected]
Abstract—This article presents the development and
implementation of a virtual tool for performing rehabilitation
exercises in persons that suffer from the carpal tunnel syndrome,
using robots (Sphero - Ollie) as haptic devices. This tool presents
different scenarios with interactive games, in which you can
perform the rehabilitation movements of the wrist of the hand,
these movements are registered by the robots, the same ones that
transmit the data of speed, position and signals of correction of
the exercises realized to the platform of virtual reality, in order to
obtain the evolution of the same. The results show the favorable
acceptance of the use of this tool as an alternative for carrying
out rehabilitation exercises in people suffering from carpal tunnel
syndrome.
Index Terms— Virtual Reality, Teaching methods, Carpal
tunnel syndrome, Rehabilitation.
I. INTRODUCTION
The study of cumulative trauma disorders that occur in
work environments as part of work routines and activities has
been a constant topic of analysis [1]. The main reasons for
disability and early retirement are caused by musculoskeletal
disorders that are produced in work environments, this type of
injuries mainly affect the muscles, tendons, nerves and blood
vessels, and that produce a great variety of physical disorders
such as contractures, strains, fiber breakage among others [2-
3], one of the most common is carpal tunnel syndrome.
Carpal tunnel syndrome, CTS, is considered a neuropathy
with a prevalence of 3% to 6% of the adult population [4];
similar studies show that unnatural postures when making
violent and irregular movements like lifting heavy loads, cause
negative effects on health, or they are also motivated by the
repetitiveness of movements and body contractions product of
forced postures when executing a certain task [5]. The Carpal
Tunnel Syndrome is a disease that originates inside the tunnel
formed by the carpus and the transverse carpal ligament
located in the wrist, as a consequence of the pressure under the
various tendons and the median nerve found there, causing
inflammation with pain, weakness and burning in the hands
and fingers, as well as tingling with numbness of the fingers.
[4], [6-7]
Treatments to this ailment, begin to make modifications in
their work routines in order that, while they perform their
work, your wrist remains in a neutral position (with the wrist
joint straight and not down), and that depending on the degree
of ailment the specialist doctor can include the application of a
splint that is used at night, anti-inflammatory medications
(corticosteroids) that relieve pain and numbness; however, if
your CTS symptoms are severe or do not improve with the
treatments mentioned, surgery should be used to free the
carpal tunnel and eliminate the pressure exerted on the median
nerve [8].
Currently, with the advance of science and technology, new
alternatives are being analyzed to heal this type of ailment.
The National Institute of Neurological Disorders and Stroke
(NINDS) is the main sponsor of the federal government of
biomedical research in neuropathy using technology as:
transcutaneous electrical stimulation, ultrasound, biofeedback,
orthosis, that have been used with patients who have CTS and
that its application has been effective [9]. Emerging
technologies such as Virtual Reality, VR, are beginning to be
applied in this type of disorders, allow to develop different
interactive scenarios capable of creating a simulation that
involves all the senses, in real time in the form of digital
images and sounds, giving the sensation of presence in this
virtual environment [6], [10-11]; there are applications that are
already being applied as is the case of the Mobile Mixed
Reality System (MoMiReS): a 3D game based on smartphones
that generates physical interactions with a wireless glove that
lead to a set of repetitive movements to rehabilitate problems
originated in the wrist of the hand, with games specifically for
the treatment of CTS, achieving high system approval ratings
[12-13].
In this context, the article proposes the development of a
virtual tool, in which interactive games are implemented in
which the rehabilitation exercises are allowed, for the CTS,
using robots that allow interaction in the proposed games with
the aim of completing rehabilitation to reduce the discomfort
caused by this syndrome.
This article is divided into 6 Sections including the
Introduction and References. Section 2 present the Problem
formulation, in which the traditional rehabilitation exercises
are described in a fast way; describes of the development of
the proposal in Section 3; in section 4 the usability analysis of
the proposed tool is presented and finally the conclusions are
detailed in Section 5.
II. PROBLEM FORMULATION
The CTS are considered as an occupational disease, some
56
authors [14] consider that the etiology of the CTS is largely
structural, genetic and biological, and that environmental and
occupational factors, such as the repetitive use of the hand,
generate this syndrome. Some of the work factors that have
been better related to the development of STC are those that
cause an increase in pressure in the carpal tunnel due to
inadequate estimation of the load on the hands. Examples of
tasks related to the STC stand out the specific position of the
hand during the performance of the task (dorsiflexion flexion,
extension and substitute), the resistance to overcome with the
fingers the grip and possession of an object, the pressure on
the hand, the repetitive movements and the work with
vibratory tools. These factors are frequently observed in the
work of people employed in meat processing, assembly of
sub-assemblies, packaging of products, or employees such as
supermarket cashiers and people who work with computers. Among the traditional exercises used as therapy to reduce the CTS are: i) Extend and stretch the wrists and fingers out, ii) Stretch both wrists forward and relax the fingers, iii) Make a frozen fist and turn to the left and iv) With your fist frozen, gently bend each wrist down. [15]
Figure 1. Extension and Stretching Exercises
Many of the traditional rehabilitation exercises can be performed on an individual basis, so that people suffering from this syndrome lose interest in carrying out rehabilitation exercises in a continuous manner, not complying with the time of the rehabilitation period indicated.
For the above, we propose a virtual tool that helps to
perform the rehabilitation exercises in a different way, through
interactive games in which these exercises can be done in an
entertaining way using robots as haptic devices that help in
carrying out the exercises and allow to record the speeds and
scopes of the movements, with which you can generate
historical evolution of the realization of them
III. PROPOSAL DESCRIPTION
The bilateral interaction between the patient and the virtual
environment is carried out through a spherical robot which is
controlled and manipulated by the patient while performing
the rehabilitation tasks determined by the virtual work
environment. The development of the virtual environment will
allow the spherical robot to emit an input signal to the virtual
environment that is analyzed to indicate whether the patient
successfully fulfilled the defined task by increasing the
complexity of the treatment, otherwise the instructions of the
rehabilitation process are repeated. In Fig. 2, the block
diagram proposed for the development of the virtual
environment is shown, which is divided into five parts.
System inputs, the input devices of the carpal tunnel
rehabilitation treatment system allow the capture of signals to
be interpreted and perform a predetermined action. The
devices used as inputs are: i) GearVR, this virtual device
allows the immersion of the patient in the virtual environment,
stimulating him to perform the rehabilitation treatment
proposed by the specialist; ii) Robot Sphero, this electronic
device allows bilateral interaction between the patient and the
virtual interface. Through the manipulation of the robot the
information is sent to fulfill the proposed rehabilitation task
while the virtual environment closes the control loop feeds the
patient through the robot Sphero the movement that must be
executed to perform the proposed task.
Figure 2. Block diagram of the virtual environment
Outputs: he system consists of electronic devices that emulate
movements, environments, sounds, among others; these output
devices are: virtual reality helmet, audio speakers and the
spherical robot that executes the movements according to the
fulfillment of the patient's rehabilitation treatment.
Virtual environment: the virtual reality environment is
developed in order to motivate the patient to carry out the
rehabilitation treatment proposed by the specialist, according
to the evolution of the patient, the complexity of the treatment
57
is increased allowing to record a record of compliance and
advances of the patient; the virtual environment was
developed on the Unity 3D platform, where it has the
respective programming of scripts that allow interacting with
the inputs and outputs of the system.
Control stage: the proposed control for the rehabilitation
treatment provides the patient's interaction in the virtual
environment through the feedback of position and speed by
means of the robot Sphero. The Sphero robot is capable of
sending and receiving signals with respect to the reference
system ( ),R X Y Fig. 3; As the patient manipulates the robot in
a coordinated and cooperative way, he will receive the
positions directly XP , YP and orientation 𝜓 allow for
monitoring in the execution of the treatment in order to
improve the rehabilitation process.
Figure 3. Reference system of Sphero Robot
Virtualization, the virtual rehabilitation environment provides
the patient with the necessary information, so that the
fulfillment of the tasks is understandable and friendly, i.e., the
environments created have a high level of immersion and
interaction with the patient. Immersion considers visual,
auditory and correction signals XF , YF , these are the ones in
charge of motivating and helping the patient in fulfilling the
tasks; while the interaction is effected through the Sphero
robot that sends the position XP , and orientation 𝜓 to the
virtual environment.
For the rehabilitation treatment of the carpal tunnel, is work in
two scenarios illustrated in Fig. 4, i) Safe box scenario,
instruction is presented that allows to open the safe for the
purpose that the patient moves from right to left the wrist; ii)
Labyrinth scenario. the exercises performed allow to stretch
the release of the pressure exerted by the median nerve.
(a) Safe box scenario
(b) Labyrinth scenario
Figure 4. Rehabilitation Scenarios
IV. METHODOLOGIC AND ANALYSIS
This section presents the methodology for the application of
interactive games using simulated instructions and actions,
using virtual reality and the Sphero robot, which are oriented
towards patients with STS depending on the intensity level of
the syndrome. The games present a series of exercises where it
is possible to control hand movement and record patient
information such as: i) intensity of movement ii) direction of
movement iii) duration of the exercise. Thus, is possible to
analyze the progress of rehabilitation, as well as the patient's
acceptability to the use of virtual environments as an
alternative treatment for STS. Depending on the specialist's
recommendations, two types of games have been proposed:
A. Game of Safe box
The game shows a safe that must be opened, entering the
correct combination, for this purpose the instructions to be
followed by the patient are indicated, i.e. position and
orientation. Each time a task is completed, the results are
displayed to record the progress of therapy. The game
interface is shown in the Fig4.
(a) Instructions
58
((b) Turn 30o to the left
(b) Turm 60o to the rigth
(c) Turm 55o to the left
(d) Error in instructions
Figure 5. Environment Game of Safe box
In this game, the patient must hold the Sphero simulating the
wheel to make movements from left to right, with many
repetitions given randomly, thus is possible to improve the
mobility of tissues that are being rehabilitated.
B. Game of Labyrinth
Applying an interactive virtual environment, it is proposed to
use the Sphero to mobilize within a labyrinth, in this game
obstacles and rewards are presented, so that in the course the
patient can be gaining points each time he catches a group of
coins, while if he collides with an obstacle will be penalized
returning to the initial location (see Fig.5).
Figure 6. Environment Game Labyrinth
While the game is running, the patient makes slight
movements from left to right and back and forth, if the patient
deviates from the predetermined path or collides with an
obstacle, the Sphero robot will send a correction signal that is
generated by means of force feedback.
Finally, to establish the interest in these games, a usability test
was carried out with two groups of people: i) patients (Qp)
who have undergone CTS treatment, and ii) physiotherapists.
For the first group, 4 CTS patients were considered who
59
combined traditional rehabilitation with two rehabilitation
sessions using the virtual tool. The second group included 3
physiotherapists, who interacted with the games so that they
could evaluate the experience in relation to the environment
and familiarization with the application. Fig. 6 shows the
results obtained for the questions posed with ten being the
highest weighting and zero the lowest weighting.
Table 1: Questions to evaluate the usability of the virtual
environment
Questions
Qp1. How familiar are you with handling devices that allow
immersion in virtual environments?
Qp2. Is the management of virtual environments easy?
Qp3. Is the execution of the games simple and intuitive?
Qp4. Are the limitations given by external noise (light, depth)
imperceptible?
Qp5. Does the equipment used cause no discomfort?
Qd1. Are patients motivated with this type of tool to perform
rehabilitation therapies?
Qd2. Does the system facilitate obtaining information about
the progress of rehabilitation?
Qd3. Does the incorporation of new games facilitate the
development of exercise routines?
Qd4 Does the system facilitate the detection of errors in
rehabilitation?
Qd5. Can the system be implemented for rehabilitation due to
similar traumas?
Figure 7. Questions results
In Fig. 6, it can be observed that both patients and
physiotherapists have acceptance for the use of this type of
exercise games to perform CTS therapies, also show the
possibility that they can be applied for the recovery of other
types of trauma.
V. CONCLUSION
In this work, a virtual system has been developed to
quantify the value of the CTS rehabilitation treatment variable
through a Sphero robot. The Sphero robot accepts linear and
angular displacements as inputs, while as outputs it provides
force and torque, which are used for system feedback
according to the type of rehabilitation movement the patient
performs.
The results obtained in the virtual environment implemented
for people suffering from STS show the efficiency of the
proposed system; the system developed was tested with
patients of different gender and age, under the supervision of a
physiotherapist who verified the validity of the movement, and
the results obtained were counteracted with traditional
therapies.
The system presents two series of virtual exercises, one in
which repetitive movements are performed to strengthen the
tissues and measure the patient's hand's ability to move; and
the other game presents a playful environment, giving the
patient greater security since it participates in the therapy as a
game and not as a therapeutic procedure, this method aims to
change the usual position of the hand or generate shocks that
help reduce pain or discomfort produced as a symptom.
REFERENCES
[1] McGaghie, W., Issenberg, B., Cohen, E., et al.: Does Simulation-based
Medical Education with Deliberate Practive Yield Better Results than Traditional Clinical Education? A Meta-Analytic Comparative Review
of the Evidence, Journal HHS Public Access, vol. 86, pp. 706-711, 2012.
[2] Sucher, B., Schreiber, A.: Carpal Tunnel Syndrome Diagnosis, Physical Medicine & Rehabilitation Clinics, vol. 25, pp. 229-247, 2014
[3] Huisstede, B., Hoogvliet, P., Franke, T., Randsdorp, M., and Koes, B.,
Carpal Tunnel Syndrome: Effectiveness of Physical Therapy and Electrophysical Modalities. An Updated Systematic Review of
Randmized Controlled Trials, Archives of Physical Medicine and
Rehabilitation, 2017 [4] Stocker, R. L., Macheiner, A.: Capitate Non-Union: One of the Causes
of Carpal Tunnel Syndrome, Handchirurgie, Mikrochirurgie, plastische
Chirurgie: Organ der Deutschsprachigen Arbeitsgemeinschaft fur Handchirurgie, vol. 48, no 3, p. 171-174, 2016.
[5] Rosales, R. S., Martin-Hidalgo, Y., Reboso-Morales, L., & Atroshi, I.: Reliability and construct validity of the Spanish version of the 6-item
CTS symptoms scale for outcomes assessment in carpal tunnel
syndrome. BMC musculoskeletal disorders, vol. 17, pp.115-125, 2016. [6] Fernández-de-las-Peñas, C., Fernández-Muñoz, J. J., Palacios-Ceña, M.,
Navarro-Pardo, E., Ambite-Quesada, S., & Salom-Moreno, J.: Direct
and indirect effects of function in associated variables such as depression and severity on pain intensity in women with carpal tunnel
syndrome, Pain Medicine, vol. 16, pp. 2405-2411, 2015.
[7] Della Fina, V., Cera, R.: Protecting the Rights of People with Autism in the Fields of Education and Employment, Ed Springer-Verlag (2015).
[8] Ennis-Cole, Demetria L.: Technology for learners with autism spectrum
disorders, Ed. Springer, 2015.
[9] Al S., Sarah, Hussain Al D.: "Descriptive characteristics of children with
autism at Autism Treatment Center, KSA, Vol. 151. Physiology &
behavior 151 pp 604-608, 2015. [10] Quevedo, W., Ortiz, J.S., Velascco, p., Sánchez, J., Álvarez, M., Rivas,
D., and Andaluz, V.H.: Assistance System for Rehabilitation and
Valuation of Motor Skills, Sugmente Reality, Virtual Reality, and Computer Graphics, Lecture Notes in Coputer Science, pp. 166-174,
2017.
[11] Huang, X., Naghdy, F., Naghdy, G., Du, H., and Todd, C.: The Combined Effects of Adaptive Control and Virtual Reality on Robot-
0
2
4
6
8
10
12
Qp1 Qp2 Qp3 Qp4 Qp5 Qd1 Qd2 Qdt3 Qd4 Qd5
Average Standard deviation
60
Assisted Fine Hand motion Rehabilitation in Chronic Stroke Patients: A
Case Stud, Journal of Stroke and Cerebrovascular Diseases, 2017
[12] Bernardini, S., Porayska-Pomsta, K., & Smith, T. J.: ECHOES: An
intelligent serious game for fostering social communication in children with autism, Vol. 264. Information Sciences, 264, pp. 41-60, 2014.
[13] Andaluz, V.H., Cartagena, P., Naranjo, J., Agreda, J., and López, S.:
Virtual Environments for Motor Fine Skills Rehabilitation with Force Feedback, Augmented reality, VirtualReality, and Computer Graphics,
Lecture Noes in computer Science, pp. 94-105, 2017
[14] Zhang, C., Li, M., Jiang, J., Zhou, Q., Xiang, L., Huang, Y., ... & Peng, W.: Diagnostic Value of Virtual Touch Tissue Imaging Quantification
for Evaluating Median Nerve Stiffness in Carpal Tunnel Syndrome,
Journal of Ultrasound in Medicine, 2017. [15] Peñas, C., Palacios, J., Fuensalida, S., Blanco, C., Pareja, J., and
Alburquerque, F.: Effectiveness of manual therapy versus sugerí in pain
processing due to carpal tunnel syndrome: A randomized clinical trial, European Journal of Pain, pp. 1266-1276, 2017.
61
Virtual Assistant of
Training Routines Physical Movements
Fernando A. Chicaiza, Christian P. Carvajal, Luis Lema-Cerda, Washington X. Quevedo, Marcelo Álvarez,
Jessica S. Ortiz, Jorge S. Sánchez and Víctor H. Andaluz Universidad de las Fuerzas Armadas ESPE, Sangolquí-Ecuador
[email protected]; [email protected]
lalema, wxquevedo, rmalvarez, jsortiz4, jssanchez, [email protected]
Abstract— This article proposes a virtual system to stimulate
the executing of physical movements routines for both, the upper
and lower limbs. The system contemplates some virtual
environments for human-computer interaction, depending on the
type of routines of exercises that the user wishes to perform, i.e.,
workout routines to lose weight, define and mark muscles, and so
on. For any selected workout routine, the interface allows
visualizing both the routine of movements that must be made by
the user, as well as the real movement executed by them. This
information is relevant for the purpose of i) stimulating the user
during the execution of movement routines, and ii) evaluating the
movements made so that the virtual personal trainer can
diagnose the evolution of the user. The visual environment
developed is based on a systematic application which the user
first analyzes and generates the necessary movements in order to
complete the defined task. The results show the efficiency of the
system generated by the human-computer interaction oriented to
the physical movement of motor skills.
Index Terms— Virtual Assistant, Virtual Reality, Training
Routines, Physical Movements.
I. INTRODUCTION
In recent years, human beings have turned to extreme
monotony in their daily activities, due to excessive comfort
provided by technology, static transportation within vehicles,
and so on, which has repercussions in extreme situations of
sedentary life [1], [2]. The World Health Organization [3]
states that one of the main causes of mortality worldwide is
the lack of physical activity, through the adoption of
cardiovascular disease, cancer and diabetes; it specifically
denotes that [4] is alleged to account for 6% of deaths
worldwide, 21% of breast cancer, 27% of colon cancer and
27% of cases of diabetes approximately. Similarly, a study
developed by the same organization [5] identifies that more
than 81% of adolescents and 23% of adults do not have a
sufficient level of physical activity and whose
recommendation stipulates at least 60 minutes a day of
different types of physical activity.
Starting from this precept, part of the population has
evolved positively towards outdoor sports in controlled
spaces, whose evolution is determined at different levels,
considering the need for mobility [6], the benefits of
cardiovascular characteristics and even the degrees of
association in search of a greater benefit that lead to an
increase in physical activity [7]-[9]. However, the lack of
time, the economic factor, the search for immediate results,
the absence of a personal trainer, eating disorders related to
balanced diets, number of meals per day and even meal times,
influence evidence towards a better physical condition.
Within the contributions of science there are several
applications, tutorial videos which can help people to develop
physical activities with better techniques[11], even within the
social communities are promoted certain miraculous
sequences that will allow the individual to identify results in
an almost immediate manner[12]. The practice of physical
activities in controlled environments, such as the home, gym
or parks, allow people to develop previously programmed
sequences of physical exercises. It is important to consider
that without the presence of a personal trainer, the risks of
injury are high; therefore, the analysis of these behaviors has
been investigated by many fields of science where studies of
systems based on fixed and mobile sensors focused on the
prognosis of human health, the impulse of VR exergames,
which allow to take advantage of virtual reality games to
create customized applications through different work
environments [13]-[20].
As described in previous paragraphs, this work proposes the
implementation of a bilateral virtual system between user-
computer applied to physical movement training routines of
upper and lower extremities. The application is developed in a
3D graphic engine which allows interaction and immersion of
the operator in friendly environments that stimulates the
execution of different training routines aimed at weight loss
and definition of muscles. The developed system allows the
user to assess the routine of exercises performed based on the
exercises proposed by the virtual personal trainer.
This article is split into 7 Sections, including Introduction
and References. Section 2 presents the structure of the
proposed virtual system; while the valuation of movements
performed by the user and the virtual training interface are
presented in Section 3 and Section 4, respectively. The
experimental results that validate the proposal are shown in
Section 5; and finally, the conclusions are detailed in Section
6.
62
II. VIRTUAL STRUCTURE SYSTEM
This paper presents the development of an application in a
3D virtual environment that offers users an alternative to
perform some assisted physical activities routines for upper
and lower extremities. In addition, the proposed application
allows routines to have a pre-evaluation of the execution of
movements during a pre-established routine; this information
will allow the personal trainer to carry out detailed studies and
analyzes of the progress of each user, and thus determine the
evolution of the user before the different physical routines of
exercises developed by the user.
The virtual personal trainer, developed in this work, allows
to have a feedback of movements through points that
represent the exoskeleton of the user, see Figure 1.
Figure 1: Skeleton tracking
This virtual system is safe and also entertains the user while
he/she is working out. Figure 2 describes the interaction of the
user with the proposed system, establishing as the main
element of the communication the visual feedback that
encompasses the two main actions within a virtual
environment, observe and act.
The interaction between the patient and the system is
established through bilateral communication, i.e., first a
graphical interface shows the movement of the exoskeleton
that the user must execute; second the user generates the
movement trying to complete the preset routine. In real time
the Kinect 2 device tracks the user's skeleton in order to
determine if the movements performed resemble those pre-
defined by the virtual personal trainer. In addition, in the
virtual environment where the user executes his physical
activities routine, the mirror effect was implemented which
through visual feedback will allow the user to correct the
movements he is executing in order to successfully perform
the defined session, i.e., the user's neuronal system becomes
in the controller of a closed-loop control system. The visual
environment developed for this type of virtual personal trainer
provides a systematic application which the user first analyzes
and generates the necessary movements in order to complete
the defined task.
NEURAL
SYSTEMVISION
MOTOR
SKILLS
UNITY 3D
(COMPUTER)
VISION
SENSOR
GRAPHICS
PATIENT MACHINE
VISUAL FEEDBACK
Figure 2: Training System block diagram
The virtual system developed is implemented on Unity 3D
graphics engine; environments in which several scenarios are
considered in order to stimulate the user's neuronal system
when executing the physical movements routines. It proposes
the valorization of the movements made by users with the
purpose of validating and diagnosing the progress the
evolution of the user before the different routines
implemented.
The programming for the operation of the scene in virtual
reality and in augmented reality graphically shown in Figure
3, which links the components with the mechanics of
movements performed by a person in front of a body tracking
device.
Figure 3: Training System Scheme
In section i) the scenes in Virtual and Augmented Reality
are developed which are chosen in the user interface at the
moment of starting the application. There are also objects that
show the data captured by the Kinect device, as well as
cameras and auxiliary objects.
63
Section ii) corresponds to SCRIPTS, where you have the
main application controller which operate all interactions and
displays information in virtual reality and augmented reality
environments by management the inputs: Kinect, Oculus
SDK, and Input / Output plugin. It uses an algorithm that
reads the positions of joints in space and overlays them to the
color image that the Kinect device captures. By means of
position comparators it is determined if the user's movements
are correct. It also has a controller to export the data acquired
from the user to databases so that external analysis can
perform analyzes that require dedicated calculation potential.
In section iii) of Outputs, the resulting responses of inputs
and interaction with the user interface are received in audio,
video and tracking outputs.
III. TASK EVALUATION
The development of the application is based on the
technology Kinect v2 that allows to measure objectively and
accurately the positions of the main joints of a human body.
The Figure 4 shows the skeleton of the user represented by 24
points, the position x - y - z of the reference points is
obtained through the function Skeleton Frame Ready with
respect to a global reference 3R< > frame located on the
Kinect device.
For the evaluation of the movements during the execution
of the test, the final positions of the reference points in the
extremities, the vectors used are; pqa for the arm ,where
:p represents the exact point of reference of s =shoulder,
e =elbow, w =wrist, h =hand, t = thumb y ht = tip of the
hand; and the position of the arm r = right or l = left, that is
the vector ( ), ,sr sr sr srx y za = would represent the position of
the right shoulder; of similar shape for the leg pql , where
:p represents the exact point of reference of h =hip,
k =knee, a =ankle, and f =foot ; and :q r = right or l =
left; for the central part of the body the references are
considered: ht for the head, sht for the central part of the
shoulders, st for the central part of the spine y ht for the
central part of the hip.
Figure 4: Representation of the assessment of the test and Tracking points
through Kinect v2
The application used is executed within the vector space of 3R< > . To measure the angles of the limbs in different
positions we use metrics defined in an Euclidean Space, for
the angle measurement the expression is used
cosi j
i j
v v
v v = , where
iv and
jv are obtained from the
difference between the coordinates of the reference points,
e.g., i sl el
v a a= − and j el wl
v a a= − , to obtain the modulus of
the vectors representing the limbs we use the definition of
norm i i i
v v v= , with these data you can evaluate each
of the tests that are required, Fig 4.
IV. VIRTUAL TRAINING INTERFACE
The application developed in Unity presents a user interface
with two profiles: i) The Trainer Gym; and ii) User, in which
the workout is tracked.
The gym trainer profile requests user input and password,
in order to contrast with the credentials stored in the database,
once access is authorized, forms are displayed that allow the
administration of information and user therapies, in which
CRUD operations (Create, Read, Update, Delete) are applied
with the corresponding records. In this profile you can find the
search option to find registered users and continue with the
appropriate physical training, the system allows you to select
from among several training levels to be applied in a virtual
way and finally present a report in which the Advances of the
person who meets the exercise routine.
2: [compare:true]
login and
password
3:
[con
ectio
n:tru
e]
Use
r inf
orm
atio
n
4:
[conectio
n:tru
e]
User
info
rmatio
n
4: [Training
given:]
Charged TrainingVirtual
System
1: [conection:true]
get login and
password
5:
[con
ectio
n:tru
e]
traki
ng
info
rmat
ion
Figure 5: Diagram of the user interface
The profile of the user asks for entering the username and
password and start the training routine, Figure 5. The
exercises can be done in an environment outside the office
and even at home if you have the necessary equipment, for
which they have defined two environments: Virtual reality
focused on people who want greater immersion in the virtual
environment by incorporating images and sounds that provide
a greater stimulus when performing the exercises, in this
64
environment you have the possibility to observe the
movements and ensure the correct realization of them; The
Augmented Reality is aimed at people who do not accept
invasive devices, in this environment the data obtained from
the Kinect is shown, which form a vectorized skeleton that
indicates the user's current position, in parallel, the user's
image it is shown in real time and the graphic of a reference
profile is superimposed with the image of the person who
performs the exercise, which allows to know the anatomical
position of each part of the body, as shown in Figure 6; the
information generated is stored in a database for further
analysis.
2: [compare:true]
login and
password
3: [T
rack
ing]
data
1:
[conection:true]
get login and
password
Figure 6: System Virtual functional diagram
The information stored in the database is processed using a
computational tool for statistical analysis, the evaluation of
the results can be shown graphically and numerically, in this
way the gym trainer can apply this information at the next
level of training.
V. RESULTS
The application results are split into two parts: a sequential
set of windows which allow the choice of various parameters
prior to the execution of an exercise and experimental tests,
both of them presented in this Section.
Interface panels: The set of sub windows are divided into
home panel, personal trainer, scenarios, and exercises, which
provide the configuration of the work environment.
Home panel: It is the main window that stores user
information (Figure 7) in order to generate historical data of
avatar type, scenario, and exercises that are frequently
selected. Otherwise, it is allowed to add information about
new users.
Figure 7: Home panel
Avatar panel: The objective of the avatar is to guide the
exercise to avoid wrong executions that may cause damage to
the performer's joints, as well as show the speed of execution
to properly develop the task. Figure 8 shows the kind of avatar
that can be selected, from which action figures and
professional trainers are presented.
Figure 8: Selectable avatar types
Scenarios: The main purpose of this selection is to
familiarize the user with the place where the exercises are
performed. The scenarios (Figure 9) are completely
virtualized so the user can have a high degree of immersion,
where they can visualize the avatar executing the exercises in
the entire virtual space. The options show a gym, a park and a
castle, the first two with people and the third option without
people.
Exercises: The exercise panel (Figure 10) is completely
varied. Tasks to exercise muscles of the upper extremities
(biceps brachial, triceps brachial, deltoids) and lower
(quadriceps, biceps fémoral, calf muscles), abdominals, grand
pectorals, back and gluteal muscles are shown by the chosen
avatar, so the user has a clear perspective on the exercise
which is going to select.
Figure 9: Selectable scenarios types
65
Figure 10: Selection of exercises
Work environment: Once the parameters for the execution
of exercises have been selected, the work environment
includes the avatar, the scenario, and the type of exercise. In
addition to this, the environment shows a screen where the
user can visualize their execution, with the aim of giving
feedback to the visual information which is lost due to the
inclusion of virtual reality glasses.
The work environment additionally displays a panel for
modifying the characteristics of the current music track, as
well as changing either the next or previous track or album
stored in an internal database of the program. The
modification for the change in the music tracks is carried out
with a set of corporal movements that the algorithm
recognizes.
Experimental tests: The results of the interface created in
virtual reality are validated with real tests developed by a
group of people. To detail the designed environment, this
section presents a group of captures of the execution of two
exercises: the first one for arms and another one for legs.
The first task (Figure 11) requires the user to exercise the
biceps muscle, a routine that is constantly being evaluated by
the Kinect sensor. The correct execution requires the user to
keep his back in a safe position while descending and
ascending a dumbbell. For the exercise to be successfully
developed, the number of repetitions and the speed of
execution must be as close as possible to the guiding avatar.
The coincidence in the execution of the exercise is evaluated
online by an internal comparison algorithm, which
accumulates the amount of time the user stays in the desired
positions. When the threshold of expected positions exceeds
an acceptable factor, the result of the task is considered
successful, otherwise, the interface shows a message of
erroneous execution.
The second task shown (Figure 12) as an example is called
dumbbell lunges, which consists of working out quadriceps,
glutes, and biceps fémoral. The internal algorithm calculates
the amount of fat burned depending on the exercise and the
user's weight. This information is included in the lower right,
allowing the user to know the evolution of the exercise.
Likewise, the training environment includes options to back
home, which allows the user to select new routines or select
another avatar as a training guide.
Remark: The whole information of the exercise, fat
burned, execution time, heart rate and other configurations are
automatically stored in a database.
a)
b)
c)
Figure 11: Execution of the first experimental test
a)
b)
c)
66
Figure 12: Execution of the second experimental test
VI. CONCLUSION
This work proposes a virtual reality application for routines
of physical movements aimed at losing weight, tone and
define muscles. The implemented system considers a sensor
for the tracking skeleton of the user, sensors for the
acquisition of bioelectric signals and virtual reality immersion
glasses. The virtual environment is configured with the user
personal data, which allows selecting the avatar in order to
stimulate the correct execution of physical movements
proposed by the virtual personal trainer. Additionally, the
system allows immersion and interaction with the virtual
environment, with the purpose of knowing the user's
bioelectric information, e.g., number of fat burned, heart rate
per minute and execution time of the exercise.
REFERENCES
[1] J. Ma et al., «Profiles of sociodemographic, behavioral,
clinical and psychosocial characteristics among primary
care patients with comorbid obesity and depression»,
Prev. Med. Rep., vol. 8, pp. 42-50, 2017.
[2] B. Kamal, «¿Do you want to prevent stroke, cancer and
diabetes? ¡Move, now!», Inter Press Service; New York,
New York, United States, 2017
[3] World Health Organization, «World health statistics
2015», WHO graphics, 2015.
[4] «OMS | Physical activity», WHO. [Online]. Available in:
http://www.who.int/mediacentre/factsheets/fs385/es/.
[Accessed: 09-nov-2017].
[5] «OMS | Worldwide recommendations about physical
activity for the health », WHO. [Online]. Available in:
http://www.who.int/dietphysicalactivity/publications/978
9241599979/es/. [Accessed: 09-nov-2017].
[6] S. B. Kritchevsky et al., «Exercise’s effect on mobility
disability in older adults with and without obesity: The
LIFE study randomized clinical trial», Obesity, vol. 25,
n.o 7, pp. 1199-1205, 2017.
[7] S. K. Agarwal, «Cardiovascular benefits of exercise», Int.
J. Gen. Med., vol. 5, pp. 541-545, 2012.
[8] B. C. Turi et al., «Association of different physical
activity domains on all-cause mortality in adults
participating in primary care in the Brazilian National
Health System: 4-Year follow-up», J. Phys. Act. Health,
vol. 14, n.o 1, pp. 45-51, 2017.
[9] C. J. Lavie, R. Arena, y S. N. Blair, «A call to increase
physical activity across the globe in the 21st century»,
Future Cardiol., vol. 12, n.o 6, pp. 605-607, 2016.
[10] S. J. Park, M. Subramaniyam, S. Hong, y D. Kim,
«Service based healthcare monitoring system for the
elderly - Physical activity and exercise», Adv. Intell. Syst.
Comput., vol. 590, pp. 337-342, 2018.
[11] D.-C. Lee, R. R. Pate, C. J. Lavie, X. Sui, T. S. Church, y
S. N. Blair, «Reply: Add 10 min for your health: The new
japanese recommendation for physical activity based on
dose-response analysis», J. Am. Coll. Cardiol., vol. 65,
n.o 11, pp. 1154-1155, 2015.
[12] J. D. Amor y C. J. James, «Validation of a Commercial
Android Smartwatch as an Activity Monitoring
Platform», IEEE J. Biomed. Health Inform., vol. PP, n.o
99, pp. 1-1, 2017.
[13] M. Fahim, A. M. Khattak, S. Aleem, y H. A. Katheeri,
«Physical activity recognizer based on multimodal
sensors in smartphone for ubiquitous-lifecare services»,
in 2017 IEEE AFRICON, 2017, pp. 524-529.
[14] Andaluz, V., Patricio, C., José, N., José, A. and Shirley,
L. «Virtual Environments for Motor Fine Skills
Rehabilitation with Force Feedback» in International
Conference on Augmented Reality, Virtual Reality and
Computer Graphics, 2018, pp. 94-105
[15] J. D. Amor y C. J. James, «Reconstructing physical
activity monitoring outcome measures using an interval
sampling approach», in 2017 39th Annual International
Conference of the IEEE Engineering in Medicine and
Biology Society (EMBC), 2017, pp. 2365-2368.
[16] Quevedo, W., Ortiz, J., Velasco, P., Sánchez, J., Álvarez
M., Rivas D., Andaluz, V. «Assistance System for
Rehabilitation and Valuation of Motor Skills» In
Augmented Reality, Virtual Reality, and Computer
Graphics, 2017, pp. 166-174.
[17] Andaluz, V., Salazar, P., Silva, M., Escudero. M.,
Bustamante, D. «Virtual Environments for Motor Fine
Skills Rehabilitation with Force Feedback» in Automatica
(ICA-ACCA), IEEE International Conference on, 2018,
22, pp. 99-104.
[18] J. Navarro et al., «Exergames on Line for Childhood
Obesity: Using a Web Platform as an Ambulatory
Program to Increase the Acceptance and Adherence to
Physical Activity (PA)», in eHealth 360°, Springer,
Cham, 2017, pp. 127-134.
[19] Andaluz, V., Salazar, P., Escudero. M., Bustamante, D.,
Silva, M., Quevedo, W., Sánchez J., Espinosa, E., Rivas,
D. «Virtual Reality Integration with Force Feedback in
Upper Limb Rehabilitation» in Advances in Visual
Computing, 2016, pp. 259-268
[20] Santana, A., Ortiz, O., Acosta, J., Andaluz, V.,
«Autonomous Assistance System for People with
Amyotrophic Lateral Sclerosis» in Lecture Notes in
Electrical Engineering, 2017, 449, pp. 267-277