Research Article Ultra Wide X-Band Microwave Imaging of ...downloads.hindawi.com/journals/ijap/2015/528103.pdf · Ultra Wide X-Band Microwave Imaging of Concealed Weapons and Explosives

Research ArticleUltra Wide X-Band Microwave Imaging of Concealed Weaponsand Explosives Using 3D-SAR Technique

P Millot and L Casadebaig

ONERA The French Aerospace Lab Electromagnetism and Radar Department (DEMR)BP74025 2 avenue E Belin 31055 Toulouse France

Correspondence should be addressed to P Millot patrickmillotonerafr

Received 19 December 2014 Revised 2 April 2015 Accepted 14 April 2015

Academic Editor Lorenzo Crocco

Copyright copy 2015 P Millot and L Casadebaig This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

In order to detect and image concealed weapons and explosives an electromagnetic imaging tool with its related signal processingis presented The aim is to penetrate clothes and to find personal-born weapons and explosives under clothes The chosen UWBfrequency range covers the whole X-band The frequency range is justified after transmission measurements of numerous clothesthat are dry or slightly wet The apparatus and the 3D near-field SAR processor are described A strategy for contour identificationis presented with results of some simulants of weapon and explosive A conclusion is drawn on the possible future of this technique

1 Introduction

The detection of concealed weapons or concealed explosiveshas now become a major issue There is a growing need tokeep secure areas in airports subways and so forthWeaponsand explosives can be carried by people in luggage and alsohidden under their clothes

Many detection techniques have been proposed for thedetection of the presence of concealed weapons Conven-tional techniques are X-rays and metal detectors Never-theless X-rays can be dangerous for human health Metaldetectors are not imager sensors and anyway cannot detectnonmetallic objects Terahertz waves are now also emergingtechniques for vision through matter as well as millimeterwave imagers [1 2]

We propose in this paper to operate in X-band (82ndash124GHz) and to use an Ultra Wide Band (UWB) frequencysweep for electromagnetic imaging It is well known thatfor an electromagnetic imaging sensor the image resolutionincreases with frequency This is why the use of the highestfrequencies seems the most interesting But this argumentconsidered alone is a bit simplistic and there is a compromiseto be found between resolution and penetration in matter

The expected benefits of using microwaves instead ofmillimeter waves are the following

(i) Wave attenuation in ldquothrough-the-clothrdquo propagationconditions increases with frequency So we expectvery low electromagnetic losseswhen crossing clothes(even when they are very thick or sweaty)

(ii) UWB microwaves are well suited for imaging anddeliver a highly resolved imageOn the contrary thereis no risk of entering into the person intimacy becauseimages do not look like optical images

(iii) There may be less ldquoclutterrdquo or ldquospecklerdquo due to rough-ness (of clothes of objects) at microwave frequencies

(iv) One can imagine building in a near future an inex-pensive tool due to low cost microwave componentsin X-band One can use also standard antennas likehorn antennas

2 Experimental Study of Transmissionthrough Clothing

Transmission measurements of microwaves through clotheshave been performed in order to assess the frequency

Hindawi Publishing CorporationInternational Journal of Antennas and PropagationVolume 2015 Article ID 528103 8 pageshttpdxdoiorg1011552015528103

2 International Journal of Antennas and Propagation

Horn antenna

Personal computer

Network analyzer

Figure 1 Microwave test bench for transmission measurements through clothing

choice To perform these measurements a network analyzer(Figure 1) is used The transmission has been measuredbetween two horn antennas (using the parameter 119878

12) For

each garment frequency measurements have been made bysweeping in frequency from 03 to 40GHz

Results are summarized in Table 1 and are given withrespect to two criteria

(i) Maximum frequency for which the attenuation isnegligible (less than 1 dB for one path)

(ii) Maximum attenuation in the 1ndash40GHz bandwidth

(Note that these results give only some magnitudes andthat attenuation versus frequency curves would show onlymeasurement errors)

As one can notice dry clothes do not cause a highattenuation for frequencies up to 40GHz The frequencyrange for an attenuation that can be totally neglected isroughly (1ndash16GHz)The effect of the presence of human sweaton underwear has also been considered of experimental wayIn order to quantify it a cotton T-shirt sprayed with saltedwater has been put under test using the same percentage asthe human sweat that is to say 35 saltThe results are givenin Figure 2

In conclusion underwear ldquowet by human sweatrdquo canpresent typically 5 dB of attenuation at frequencies exceeding10GHz and so 10 dB for the round tripThis valuemay remainacceptable depending on the sensitivity of the final detector

3 UWB Microwave Imaging Tool Hardware

In order to estimate the shape of the desired object it isnecessary to perform three-dimensional microwave imagingAs one knows three-dimensional imaging is the best solution

Table 1 Results on transmission tests on clothes

Garments

Max frequencyfor whichattenuationlt1 dB

Maxattenuation inthe whole

1ndash40GHz band1 leather jacket (100leather) 20GHz 1 dB

2 cotton shirt (100cotton) 16GHz 2 dB

3 thick sweater (50 wool50 acrylic) 20GHz 1 dB

4 lightweight wool jacket(100 wool) 20GHz 1 dB

5 shirt (80 cotton and20 polyester) 20GHz 1 dB

6 parka (waterproofcanvas filled) 17GHz 2 dB

7 wool jacket (100 wool) 165 GHz 1 dB8 waterproof jacket (innerside 100 cotton outerside 100 polyester)

16GHz 2 dB

in terms of object identification because the sizes (in 119883 119884)and ldquodepthrdquo (in 119885 parallel to the bore sight) of the object canbe retrieved (see Figure 3)

A bidimensional synthetic antenna is required for theimaging process The best solution for real-time measure-ments would be to build an UWB bidimensional array ofantennas (see Figure 4 on the left) When the number ofantennas is large real-time operations can be obtained byantenna switching within an antenna array

International Journal of Antennas and Propagation 3

0 5 10 15 200

1

2

3

4

5

6

7

8

9

10

Frequency (GHz)

Leve

l (dB

)

Through cloth attenuation wet T-shirt

Figure 2 Attenuation versus frequency for a wet T-shirt

However in the framework of this research the solutionof the displacement of a single antenna that simulates asynthetic aperture has been adopted This simple techniqueof scanning a plane surface simulates an array of antennas Infact only one antenna is needed and the microwave device isquite simple (see Figure 4 on the right)

4 Near-Field Imaging Using3D-SAR Algorithm

Inverse scattering considering the 3D nature is a promisingfield for dielectric objet identification [3] For this purposemany algorithms in the near-field configuration have beendeveloped [4 5] For microwave imaging we have used SAR(synthetic aperture radar) imaging Becausewe have achievedthe experimental conditions of the antenna synthesis UWBSAR imaging has become very performing The idea is toadopt algorithms that are fast and robust to noise and toartifacts coming from the standing waves

SAR imaging is based upon back-propagation principleand is analogous to Kirchhoff rsquos migration Due to the lengthof the synthetic antenna and the distance to the object (40 cmfor both) we are in near field of the synthetic antenna andso special algorithms derived from spot-light SAR or seismicmigration can be used Basically the radar image is built bysumming delayed signals acquired for a very large numberof antennas and frequencies The frequency formulation isbased on phase shifts compensations [6]The 2D algorithm isalso explained in [7] and an extension to 3D is now proposed

Let 119878119903be the complex signal recorded for antenna points

(119909119894 119910119897) (number of 119873

119909119873119910) for a set of 119873

119891frequencies Then

the local reflectivity 120574 of the target at the point119872(119909 119910 119911) canbe computed as

120574 (119909 119910 119911) =

10038161003816100381610038161003816100381610038161003816100381610038161003816

120572

119873119909

sum

119894=1

119873119910

sum

119897=1

119873119865

sum

119896=1

119882(119909119894 119910119897 119909 119910 119891

119896) 119878lowast

119903(119909119894 119910119897119891119896) 1198892(119909 minus 119909

119894 119910 minus 119910

119896 119911) 119890minus119895(4120587119891

119896119888)119889(119909minus119909

119894119910minus119910119896119911)

10038161003816100381610038161003816100381610038161003816100381610038161003816

(1)

where 120572 is a proportionality factor depending on normaliza-tion 119909

119894 119910119897are the coordinates of the antenna points (total

number119873119909lowast119873119910) 119891119896is the current frequency (total number

119873119865)119889 is the distance fromantenna to the current point (119909 119910)

and 119878lowast119903denotes the complex conjugate of 119878

119903

119882 is supposed to be a 3D weight function to achieveapodization and antenna gain deconvolution

119882(119909119894 119910119897 119909 119910 119891

119896) =

Ham3 (119891 119909119894 119910119897)

119891119896119866 (119909119894 119910119897 119909 119910 119891

119896) (2)

Ham3 designs the 3D Hamming functions in frequencyand space used to reduce side-lobes in the 3 dimensions(2) and 119866 is the antenna gain factor The chosen antennais a pyramidal horn of 15 dB gain and of 30-degree 3 dBbeamwidth In practice it has been found that the weightfunction along 119909 and 119910 is unnecessary as the antenna patternfurnishes a kind of natural apodization and the frequencyscaling is a smooth function The 1198892 multiplication in (1) isalso quite ineffective as the distance does not vary rapidly onthe contrary of the phase term

The space sampling requirement is at least 1205824 for SARprocessing Frequency sampling must also respect the crite-rion of antialiasing in range domain This leads to a hundredof frequency samples Typical figures are then 100 antennapositions and 100 frequency steps This is why the basic

algorithm of (1) using three sums has a computation time ofaround 10 minutes for a cube of 05 times 05 times 05 = 0125m3In order to improve the computation efficiency we havedeveloped a new algorithm with another formulation Wehave thought to use the 2D Fast Fourier Transform (FFT

2)

algorithm for computation [8] Let us construct 119872 as thematrix of measured data of dimensions119873

119909lowast119873119910lowast119873119891 In fact

squaredmatrices of dimensions119873119886

2 (number of antenna datapoints) are stacked along the frequency dimension in order tocreate the large matrix119872 Then a ldquofocusing operatorrdquo Foc iscomputed knowing the geometry of the radar scene (119909 119910 119911)and the synthetic aperture (119909

119886 119910119886) It is computed once and

then stored for all The existence of such an operator is dueto the formalism of the double convolution in 119909 and 119910 forthe focusing processing This means that the local reflectivity120574(119909 119910 119911) instead of (1) can be expressed as

120574 (119909 119910 119911) = sum

119891

[119872 (119909 119909119886 119910 119910119886 119911 119891)

lowast lowastFoc (119909 119909119886 119910 119910119886 119911 119891)]

= sum

119891

[FFTminus12[FFT2[119872 (119909 119909

119886 119910 119910119886 119911 119891)]

sdot FFT2[Foc (119909 119909

119886 119910 119910119886 119911 119891)]]]

(3)


Y

Z

X

Figure 3 Target geometry

Using the 2D-FFT (Two-Dimensional Fast Fourier Trans-form) algorithm for the Fourier Transform in two dimensionsover the (119909 119910) coordinates the imaging algorithm (see (3))became very efficient The 3D-SAR computation time for apixel of 1 cm and a volume of 0125m3 with 119873

119886= 51 and

119873119891= 201 was about 2 seconds in MATLAB 8 on a standard

PC

5 Experimental Results

The ldquoradar toolrdquo is a versatile microwave imaging devicewith very high spatial resolution due to the combinationof Ultra Wide Bandwidth and large synthetic apertureThe microwave electronics is based on a portable NetworkVector Analyzer (NVA) used both as transmitter and asreceiver It is linked to a PC equipped with the MATLAB8 software The chosen configuration for the measurementis the 119878

11reflection parameter in frequency domain Thus

only one antenna is used for transmitting and receiving Thisis to avoid bistatic angular shifts in case of two separatedantennas operating in near fieldThe bidimensional syntheticaperture (119883119884) is synthetized with the help of a two-axistranslator (Figure 5) The transmitted power is 1mW Thespace sampling is 75mm which corresponds to a quarter ofthe mean carrier wavelength (3 cm at 10GHz) The overalldimensions for the synthetic aperture in both directions are375mm and 375mmThis value has been carefully selected asa compromise between image resolution and measurementtime This means that the real image resolution is roughly3 cm in 119883 and 119884 direction at 40 cm of distance between theantenna and the target Along119885 the resolution is fixed by thefrequency bandwidth and is about 4 cm

Table 2 lists the specifications of ldquoMIXIMrdquo themicrowaveimager

The three chosen objects for the test measurement aredescribed in Table 3

The measurements are made with a different object ateach time The objects are suspended between the coat (item6) and an electromagnetic flat absorber that simulates thehuman body As it is not possible to take a human being

Table 2 Specifications of the microwave imager

MIXIM (microwave X-band imager)Operating frequency 82ndash124GHzNumber of frequency samples 201Transmitted power 1mW119879119909119877119909antenna Pyramidal horn with 15 dB gain

Video bandwidth 30KHzSynthetic aperture length 375mmNumber of antenna samples 2601119909-axis step 75mm119910-axis step 75mmTotal measurement time 24minutes119883 119884 resolution 3 cm119885 resolution 4 cm

because of the measurement time we simulate like this thewave absorption by the human being and the reflection onits surface could be similar to that of human fat The distancebetween the antenna and the object is 40 cm in order to keepthe nominal resolution of around 3 cm (see Figure 6)

The imaging results are now presented First we haveobtained 3D images using the signal processing algorithm 3Dimages can be seen as a set of 2D images (or cuts) for instancetaken in the vertical plane One can choose a 2D image in thevertical plane at random But we have adopted the followingbetter strategy in order to detect the object we remove themean value of each 2D image and we keep the pixels that areabove this value Then we take 10 vertical cuts separated by1 cmaround the range of 40 cmThese 2D images are summedin intensity normalized to themaximumvalue and displayedwith a color code in intensity between 30 and 100 of thestrongest pixel

The first image (Figure 7) represents the ceramic knifeunder cloth 6Thehandle ismetal (on the left) and the bladeis ceramic (on the right) The strongest local reflectivity is onthemetal (note that the handle is cambered) but the dielectricblade is clearly visible because it is flatThe dielectric constantof ceramic is estimated to 9

The second object is a plastic toy gun under the samecloth FromFigure 8 one can clearly imagine the shape of thisobject

The third object is the sand pouch in vertical positionWehave selected very fine sand coming from a beach in orderto simulate an explosive mass Its dielectric constant is 26FromFigure 9 one can estimate the cross section of explosivemassWhen the explosive is thick enough (say 10 cm) we canestimate its thickness and so the volume Of course explosiveappears here only as a large ldquoanomalyrdquo which is enough totrigger an alert

For objects of 4 cm size the imager will deliver just aspot which can be satisfactory for detection but not for shapeidentification This is why we claim that the dimensionsshould be at least 10 cm for better shape estimation

An extra measurement has been made for the case ofthe ceramic knife under a wet shirt (slightly watered witha spray of salted water in order to simulate human sweat)


array2D antenna

(a)

antennaHorn

(b)

Figure 4 3D EM imaging technical solutions on the right single antenna on a 2-axis translator on the left a fixed 2D antenna array

D D

Axis controller

NVAPC

Dual-axis translator

Horn antenna

Jacket

EM absorber foam

Figure 5 Schematic diagram of the imaging tool in operation

The whole detection remains good One can see that themetallic handle disappears a little but not the blade Thereason is the following Due to the amount of water containedin the shirt the shirt provides a large electromagnetic back-scattering of large extent in the (119883119884) plane As the metal isin contact with the wet shirt the processing that subtractsthe mean value is partially confused The blade that is not incontact with the shirt appears clearly on the image due to the119885-axis resolution capabilities of 3D imaging (see Figure 10)

6 Randomly Sparse Synthetic Array Imaging

In order to decrease the very longmeasurement time one canimagine synthetizing a large sparse array [9] The idea is toskip antennameasurements points along119883 and119884 to go fasterA configuration has been found with a random distributionof elementary antennas This configuration needs about 10only of the full number of antenna pointsThe selection of theconfiguration is made by the following procedure


Table 3 List description and photographs of imaged objects

Case number Object Material Overalldimensions Photographs

1 Knife Metalceramic 4 cm times 30 cm

2 Toy gun Plastic 8 cm times 19 cm

3 Sand pouch Sand 7 cm times 20 cm

Radar

Figure 6 Configuration of the measurement (object is maintainedbetween the coat and the absorber)

(i) Create a sparse matrix (ldquomaskrdquo) of 0 and 1 of dimen-sions 51 lowast 51 with 119909 sparsity

(ii) Compute the response of a point-like target by simu-lation at a distance of 40 cm in near field

(iii) Compute the response of the imaging processing byturning on and off the elementary radiators followingthe random rule given by the mask

(iv) Adopt it according to a criterion of artifact level (egminus15 dB of energy under the maximum peak)

With 119909 = 10 sparsity one can already find goodrepresentations In this case amask is represented in Figure 11and its image is given in Figure 12 (corresponding to the caseof Figure 8)

(m)

(m)

Knife

minus015 minus01 minus005 0 005 01 015

minus015

minus01

minus005

0

005

01

015

Figure 7 Ceramicmetal knife under coat (horizontal)

7 Conclusion

Short range UWBmicrowave imaging has been developed inX-band and has proved its interest It enables vision throughclothes of objects of more than roughly 10 cm dimensionsShapes can be estimated even though the image quality is notthe same as optics or THz because of the wavelength Theinteresting point is that we have found that the transmission


Toy gun


minus015

minus01

minus005

0

005

01

015

(m)

(m)

Figure 8 Plastic toy gun (horizontal)

Sand pouch

minus015

minus01

minus005

0

005

01

015

minus015 minus01 minus005 0 005 01 015(m)

(m)

Figure 9 Sand pouch under coat (vertical)

Knife under wet shirt

minus015

minus01

minus005

0

005

01

015


(m)

Figure 10 Knife under wet shirt (vertical)

5

10

15

20

25

30

35

40

45

50

5 10 15 20 25 30 35 40 45 50

Figure 11 Sparse antenna array configuration in black (1 = on) inwhite (0 = off)

Toy gun sparse

minus015

minus01

minus005

0

005

01

015


(m)

Figure 12 Toy gun with very sparse configuration

through clothes even slightly wet does not disturb theimages and so inexpensive microwave systems could beenvisagedThemain drawback comes from themeasurementtime At this time it is not possible to make measurementswith a real person Signal processing computation time is nomore a problem with 3D-SAR clever formulations In orderto improve the measurement time a research axis has beenopened Sparse array configurations already give good results(even with some unwanted side-lobes) and the researches areongoing in the field of electromagnetic imaging algorithmsThree new ideas are emerging

(i) The first one is technical it consists in using an arcwith antennas (equipped with digital switches) andto sweep the human body along a circle as a scannerFor the purpose a line of antennas (like open wave-guides) can been used as well as miniaturized FMCWradar heads


(ii) The second one is the MIMO-SAR or sparse config-urationmdashboth on transmission and on receptionmdashfor the antenna array which may bring about anotherdrastic decreasing of the number of measurements atthe price of a reduced signal-to-noise ratio

(iii) Last microwave imaging at mm wave will also beperformed in order to fuse information between X-band and mm wave (60GHz)

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

Theauthors are very grateful to the EuropeanDefenseAgencyfor having partially funded the ONERA Lab for explosivedetection studies in the framework of the E-STAR project

References

[1] DM SheenD LMcMakin andT EHall ldquoThree-dimensionalmillimeter-wave imaging for concealed weapon detectionrdquoIEEE Transactions onMicrowaveTheory and Techniques vol 49no 9 pp 1581ndash1592 2001

[2] R Appleby and H B Wallace ldquoStandoff detection of weaponsand contraband in the 100GHz to 1 THz regionrdquo IEEE Trans-actions on Antennas and Propagation vol 55 no 11 pp 2944ndash2956 2007

[3] G Gennarelli I Catapano F Soldovieri and R Persico ldquoOnthe achievable imaging performance in full 3-D linear inversescatteringrdquo IEEE Transactions on Antennas and Propagationvol 63 no 3 pp 1150ndash1155 2015

[4] R Solimene A Brancaccio R Di Napoli and R Pierri ldquo3Dsliced tomographic inverse scattering experimental resultsrdquoProgress in Electromagnetics Research vol 105 pp 1ndash13 2010

[5] A H Golnabi P M Meaney N R Epstein and K D PaulsenldquoMicrowave imaging for breast cancer detection advances inthree-dimensional image reconstructionrdquo in Proceedings of the33rd Annual International Conference of the IEEE Engineeringin Medicine and Biology Society pp 5730ndash5733 Boston MassUSA August-September 2011

[6] S E Assad I Lakkis and J Saillard ldquoHolographic SARimage formation by coherent summation of impulse responsederivativesrdquo IEEE Transactions on Antennas and Propagationvol 41 no 5 pp 620ndash624 1993

[7] M Soumekh Synthetic Aperture Radar Signal Processing JohnWiley amp Sons 1999

[8] J Fortuny and A J Sieber ldquoFast algorithm for a near-field syn-thetic aperture radar processorrdquo IEEE Transactions on Antennasand Propagation vol 42 no 10 pp 1458ndash1460 1994

[9] Y Qi Y Wang X Peng W Tan and W Hong ldquoApplication ofoptimized sparse antenna array in near range 3D microwaveimagingrdquo in Proceedings of the 17th International Symposiumon Antennas and Propagation (ISAPrsquo12) pp 1397ndash1400 NagoyaJapan November 2012

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



Horn antenna

Personal computer

Network analyzer

Figure 1 Microwave test bench for transmission measurements through clothing

choice To perform these measurements a network analyzer(Figure 1) is used The transmission has been measuredbetween two horn antennas (using the parameter 119878

12) For

each garment frequency measurements have been made bysweeping in frequency from 03 to 40GHz

Results are summarized in Table 1 and are given withrespect to two criteria

(i) Maximum frequency for which the attenuation isnegligible (less than 1 dB for one path)

(ii) Maximum attenuation in the 1ndash40GHz bandwidth

(Note that these results give only some magnitudes andthat attenuation versus frequency curves would show onlymeasurement errors)

As one can notice dry clothes do not cause a highattenuation for frequencies up to 40GHz The frequencyrange for an attenuation that can be totally neglected isroughly (1ndash16GHz)The effect of the presence of human sweaton underwear has also been considered of experimental wayIn order to quantify it a cotton T-shirt sprayed with saltedwater has been put under test using the same percentage asthe human sweat that is to say 35 saltThe results are givenin Figure 2

In conclusion underwear ldquowet by human sweatrdquo canpresent typically 5 dB of attenuation at frequencies exceeding10GHz and so 10 dB for the round tripThis valuemay remainacceptable depending on the sensitivity of the final detector

3 UWB Microwave Imaging Tool Hardware

In order to estimate the shape of the desired object it isnecessary to perform three-dimensional microwave imagingAs one knows three-dimensional imaging is the best solution

Table 1 Results on transmission tests on clothes

Garments

Max frequencyfor whichattenuationlt1 dB

Maxattenuation inthe whole

1ndash40GHz band1 leather jacket (100leather) 20GHz 1 dB

2 cotton shirt (100cotton) 16GHz 2 dB

3 thick sweater (50 wool50 acrylic) 20GHz 1 dB

4 lightweight wool jacket(100 wool) 20GHz 1 dB

5 shirt (80 cotton and20 polyester) 20GHz 1 dB

6 parka (waterproofcanvas filled) 17GHz 2 dB

7 wool jacket (100 wool) 165 GHz 1 dB8 waterproof jacket (innerside 100 cotton outerside 100 polyester)

16GHz 2 dB

in terms of object identification because the sizes (in 119883 119884)and ldquodepthrdquo (in 119885 parallel to the bore sight) of the object canbe retrieved (see Figure 3)

A bidimensional synthetic antenna is required for theimaging process The best solution for real-time measure-ments would be to build an UWB bidimensional array ofantennas (see Figure 4 on the left) When the number ofantennas is large real-time operations can be obtained byantenna switching within an antenna array


0 5 10 15 200

1

2

3

4

5

6

7

8

9

10

Frequency (GHz)

Leve

l (dB

)








(119909119894 119910119897) (number of 119873

119909119873119910) for a set of 119873



120574 (119909 119910 119911) =

10038161003816100381610038161003816100381610038161003816100381610038161003816

120572

119873119909

sum

119894=1

119873119910

sum

119897=1

119873119865

sum

119896=1

119882(119909119894 119910119897 119909 119910 119891

119896) 119878lowast

119903(119909119894 119910119897119891119896) 1198892(119909 minus 119909

119894 119910 minus 119910

119896 119911) 119890minus119895(4120587119891

119896119888)119889(119909minus119909

119894119910minus119910119896119911)

10038161003816100381610038161003816100381610038161003816100381610038161003816

(1)






119903


119882(119909119894 119910119897 119909 119910 119891

119896) =

Ham3 (119891 119909119894 119910119897)

119891119896119866 (119909119894 119910119897 119909 119910 119891

119896) (2)




2)







120574 (119909 119910 119911) = sum

119891

[119872 (119909 119909119886 119910 119910119886 119911 119891)


= sum

119891

[FFTminus12[FFT2[119872 (119909 119909

119886 119910 119910119886 119911 119891)]

sdot FFT2[Foc (119909 119909

119886 119910 119910119886 119911 119891)]]]

(3)


Y

Z

X



119886= 51 and


PC



















array2D antenna

(a)

antennaHorn

(b)


D D

Axis controller

NVAPC


Horn antenna

Jacket

EM absorber foam











Radar







(m)

(m)

Knife


minus015

minus01

minus005

0

005

01

015


7 Conclusion



Toy gun


minus015

minus01

minus005

0

005

01

015

(m)

(m)


Sand pouch

minus015

minus01

minus005

0

005

01

015


(m)



minus015

minus01

minus005

0

005

01

015


(m)


5

10

15

20

25

30

35

40

45

50

5 10 15 20 25 30 35 40 45 50


Toy gun sparse

minus015

minus01

minus005

0

005

01

015


(m)









Acknowledgment


References












RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0 5 10 15 200

1

2

3

4

5

6

7

8

9

10

Frequency (GHz)

Leve

l (dB

)








(119909119894 119910119897) (number of 119873

119909119873119910) for a set of 119873



120574 (119909 119910 119911) =

10038161003816100381610038161003816100381610038161003816100381610038161003816

120572

119873119909

sum

119894=1

119873119910

sum

119897=1

119873119865

sum

119896=1

119882(119909119894 119910119897 119909 119910 119891

119896) 119878lowast

119903(119909119894 119910119897119891119896) 1198892(119909 minus 119909

119894 119910 minus 119910

119896 119911) 119890minus119895(4120587119891

119896119888)119889(119909minus119909

119894119910minus119910119896119911)

10038161003816100381610038161003816100381610038161003816100381610038161003816

(1)






119903


119882(119909119894 119910119897 119909 119910 119891

119896) =

Ham3 (119891 119909119894 119910119897)

119891119896119866 (119909119894 119910119897 119909 119910 119891

119896) (2)




2)







120574 (119909 119910 119911) = sum

119891

[119872 (119909 119909119886 119910 119910119886 119911 119891)


= sum

119891

[FFTminus12[FFT2[119872 (119909 119909

119886 119910 119910119886 119911 119891)]

sdot FFT2[Foc (119909 119909

119886 119910 119910119886 119911 119891)]]]

(3)


Y

Z

X



119886= 51 and


PC



















array2D antenna

(a)

antennaHorn

(b)


D D

Axis controller

NVAPC


Horn antenna

Jacket

EM absorber foam











Radar







(m)

(m)

Knife


minus015

minus01

minus005

0

005

01

015


7 Conclusion



Toy gun


minus015

minus01

minus005

0

005

01

015

(m)

(m)


Sand pouch

minus015

minus01

minus005

0

005

01

015


(m)



minus015

minus01

minus005

0

005

01

015


(m)


5

10

15

20

25

30

35

40

45

50

5 10 15 20 25 30 35 40 45 50


Toy gun sparse

minus015

minus01

minus005

0

005

01

015


(m)









Acknowledgment


References












RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Y

Z

X



119886= 51 and


PC



















array2D antenna

(a)

antennaHorn

(b)


D D

Axis controller

NVAPC


Horn antenna

Jacket

EM absorber foam











Radar







(m)

(m)

Knife


minus015

minus01

minus005

0

005

01

015


7 Conclusion



Toy gun


minus015

minus01

minus005

0

005

01

015

(m)

(m)


Sand pouch

minus015

minus01

minus005

0

005

01

015


(m)



minus015

minus01

minus005

0

005

01

015


(m)


5

10

15

20

25

30

35

40

45

50

5 10 15 20 25 30 35 40 45 50


Toy gun sparse

minus015

minus01

minus005

0

005

01

015


(m)









Acknowledgment


References












RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










array2D antenna

(a)

antennaHorn

(b)


D D

Axis controller

NVAPC


Horn antenna

Jacket

EM absorber foam











Radar







(m)

(m)

Knife


minus015

minus01

minus005

0

005

01

015


7 Conclusion



Toy gun


minus015

minus01

minus005

0

005

01

015

(m)

(m)


Sand pouch

minus015

minus01

minus005

0

005

01

015


(m)



minus015

minus01

minus005

0

005

01

015


(m)


5

10

15

20

25

30

35

40

45

50

5 10 15 20 25 30 35 40 45 50


Toy gun sparse

minus015

minus01

minus005

0

005

01

015


(m)









Acknowledgment


References












RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation















Radar







(m)

(m)

Knife


minus015

minus01

minus005

0

005

01

015


7 Conclusion



Toy gun


minus015

minus01

minus005

0

005

01

015

(m)

(m)


Sand pouch

minus015

minus01

minus005

0

005

01

015


(m)



minus015

minus01

minus005

0

005

01

015


(m)


5

10

15

20

25

30

35

40

45

50

5 10 15 20 25 30 35 40 45 50


Toy gun sparse

minus015

minus01

minus005

0

005

01

015


(m)









Acknowledgment


References












RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Toy gun


minus015

minus01

minus005

0

005

01

015

(m)

(m)


Sand pouch

minus015

minus01

minus005

0

005

01

015


(m)



minus015

minus01

minus005

0

005

01

015


(m)


5

10

15

20

25

30

35

40

45

50

5 10 15 20 25 30 35 40 45 50


Toy gun sparse

minus015

minus01

minus005

0

005

01

015


(m)









Acknowledgment


References












RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation














Acknowledgment


References












RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









Documents

Research Article Ultra Wide X-Band Microwave Imaging of ...downloads.hindawi.com/journals/ijap/2015/528103.pdf · Ultra Wide X-Band Microwave Imaging of Concealed Weapons and Explosives