Embed Size (px)
Citation preview
Int. J. Electron. Commun. (AEÜ) 70 (2016) 1219–1227
Contents lists available at ScienceDirect
International Journal of Electronics andCommunications (AEÜ)
journal homepage: www.elsevier .com/locate /aeue
Regular paper
Optimal implementation of novel WCDMA uplink outer loop powercontrol algorithm
http://dx.doi.org/10.1016/j.aeue.2016.06.0071434-8411/� 2016 The Author(s). Published by Elsevier GmbH.This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
⇑ Corresponding author.E-mail addresses: [email protected] (A.G. Markoc), [email protected]
(G. Šišul).
Angelo Gary Markoc a,⇑, Gordan Šišul b
aRadio Access Network Development Department, Vipnet d.o.o., Vrtni put 1, 10000 Zagreb, CroatiabDepartment of Wireless Communications, Faculty of Electrical Engineering and Computing, Unska 3, 10000 Zagreb, Croatia
a r t i c l e i n f o
Article history:Received 25 January 2016Accepted 10 June 2016
Keywords:Performance management statisticWCDMAsirMaxUL RSSIGPEH
a b s t r a c t
Wideband code division multiple access (WCDMA) uplink outer loop power control has been responsiblefor changing the signal to interference ratio (SIR) target value according to the SIR target control algo-rithms. Thus maintains the desired block error rate (BLER) for voice connections and transmission to tar-get error rate (TTE) for data connections. During a call, the SIR target varies between implemented lower(sirMin) and upper boundaries (sirMax), which are fixed values on radio network control station (RNC).As possible solution for WCDMA uplink overload problems, a novel uplink outer loop power control algo-rithm with dynamic change of sirMax has been proposed. In this paper optimal implementation fordynamic change of sirMax has been investigated. Performance management statistic and general perfor-mance event handling (GPEH) data, for current uplink received signal strength indicator (UL RSSI) andrequired SIR target, have been obtained. Based on analysis from obtained data optimal implementationfor dynamic change of sirMax has been proposed.� 2016 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction
Power control is essential in WCDMA systems to provide satis-factory quality of service (QoS) [1] and to combat severalproblems:
� To mitigate the fading effect.� To adjust the radiated power of mobile stations in a way that allreceived signals at the base station have the same SIR for thesame bit rate [2].
� To reduce co-channel interference by concurrent users.
SIR in the WCDMA system corresponds to the ratio of thereceived useful signal at the base station receiver over a frequencyrange of 5 MHz width compared to interfering co-channel sources.In the uplink, received useful signal at the base station receiver issignal from the mobile station. Sources of interfering signals areall other mobile stations in the cell, as well as other mobile stationsin neighboring WCDMA cells that transmit on the same channel[3]. There are three types of algorithm for power control, investi-gated in [4–10], which are implemented in WCDMA system: open
loop power control, inner loop power control and outer loop powercontrol. Based on analysis of uplink air interface overload prob-lems, it has been concluded that existing uplink outer loop powercontrol has difficulty to combat uplink overload in real networks.Thus self-optimizing uplink outer loop power control withdynamic change of sirMax, based on current uplink load, has beenproposed [11]:
sirMax½dB� ¼ MaxULRSSI½dBm� � ULRSSI½dBm� þ C½dB� ð1Þ
where sirMax is value used by uplink outer loop power control,MaxULRSSI is maximum allowed UL RSSI of the cell, ULRSSI is mea-sured UL RSSI of the cell and C is constant that is used by operator tomodify dependency of sirMax to measured UL RSSI. During a call,the SIR target varies between implemented lower (sirMin) andupper boundaries (sirMax), which are fixed values on RNC in exist-ing uplink outer loop power control. On the other hand, in self-optimizing uplink outer loop power control upper boundary (sir-Max) has been load dependent. The comparison of two outer looppower control algorithms has been presented in next section.
2. Comparison of uplink outer loop power control algorithms
Cooperation between three types of uplink power control algo-rithms has been presented at Fig. 1.
Fig. 1. Cooperation between three types of uplink power control algorithms.
1220 A.G. Markoc, G. Šišul / Int. J. Electron. Commun. (AEÜ) 70 (2016) 1219–1227
Open loop power control is used to calculate the required radi-ated power by the mobile station for access preamble, the randomaccess channel (RACH) and the initial power for dedicated channels(DCH). Once dedicated channels have been established outer andinner loop power control shall cooperate in order to maintain thenecessary block error rate (BLER) for voice or transmission targeterror rate (TTE) for data calls. This is done in a way that outer looppower control sets and adjusts the SIR target value used by innerloop power control. Outer loop power control monitors the cyclicredundancy check (CRC) of transport blocks after diversity combin-ing at the control station and change the SIR target value accordingto the SIR target control algorithms for the uplink. Thus maintainsthe desired BLER and TTE in the uplink, regardless of user’s radioconditions, and whether user is stationary or mobile. The upperboundary for SIR target value is determined by sirMax parameteron RNC, which is load dependent value in self-optimizing uplinkouter loop power control algorithm presented at Fig. 2.
By implementation of load dependent sirMax value, maximumallowed radiated power per mobile station has been dynamicallyadapted to current uplink load, which could bring improvementin provided QoS during uplink overload. On the other hand, duringnormal uplink load sirMax would be increased, thus allowinghigher radiated power from mobile station and better uplinkthroughput. Dynamic changes of sirMax have been strongly depen-dent of MaxULRSSI setting that defines targeted maximum allowedUL RSSI of the cell, set by operator. For measured UL RSSI aboveMaxULRSSI, WCDMA cell is considered to be in a state of uplinkoverload, thus threshold for uplink overload has been presentedin next section.
3. WCDMA uplink overload threshold
The Shannon–Hartley theorem states that the channel capacityis given by:
C ¼ Blog2ð1þ S=NÞ ð2Þ
where C is the capacity in bits per second, B is the bandwidth of thechannel in Hertz, and S/N is the signal to noise ratio (SNR). In thispaper the focus is on uplink interference limited scenario, thusSNR would be the most relevant measure to determine WCDMAuplink overload threshold. On the other hand, uplink air interfaceutilization measurement techniques on RNC are based on UL RSSImeasurements. Thus UL RSSI has been used, instead of SNR, todetermine WCDMA uplink overload threshold. Maximum allowedUL RSSI of the cell has been defined as maximum allowed noise riseover thermal noise due to interference at which it is still possible todistinguish users [12].
UL RSSImax ¼ UL RSSI0 þ NRmax ð3Þ
where UL RSSImax is maximum allowed UL RSSI of the cell, UL RSSI0is measured UL RSSI of empty cell and NRmax is maximum allowednoise rise. If there are no users in the cell, and if cell has not beeninterfered by transmitting mobile stations from neighboring basestations, the measured signal strength at the base station receiver(UL RSSI0) correspond to thermal noise [13]. Thermal noise levelat the base station receiver can be estimated by techniquesdescribed in [14], which gives UL RSSI0 equal to:
UL RSSI0 ¼ �105 dBm ð4ÞEq. (5) from [15] has been used to calculate noise rise, which is
equal to:
NR ¼ �10� logð1� gULÞ ½dB� ð5Þ
where gUL is the uplink load factor. Statistical analysis from live net-work gives result that satisfying call setup success rate (CSSR) canbe achieved with gUL ¼ 0:98, due to admission and congestion pro-cedures specified in [16]. From (5), calculated maximum noise rise(NRmax) is then 16.99 dB. From (3), UL RSSImax value is then�88.01dBm. Thus WCDMA cells with measured UL RSSI above�88.01dBm have been considered as cells in state of uplinkoverload.
Fig. 2. Cooperation between three types of uplink power control with proposed changes for outer loop power control.
A.G. Markoc, G. Šišul / Int. J. Electron. Commun. (AEÜ) 70 (2016) 1219–1227 1221
Conditions from live network that could produce uplink over-load are divided in 3 groups:
� Uplink air interface overload due to large number of simultane-ous users, like concerts, sport events, traffic jams on highwayetc. [17].
� Uplink air interface overload caused from outer source of inter-ference [18]. Sources of outer interference could be malfunc-tioning radio devices (TV antenna amplifier, DECT systems,FAX devices etc.) or mobile operators from neighboring coun-tries that use same frequency spectrum.
� Uplink air interface overload caused by passive inter modula-tion products created by own system, which has become a sig-nificant factor with increased spectrum bandwidth acquired byoperators [19].
Optimal implementation for dynamic change of sirMax has beeninvestigated based on statistical data from performance manage-ment statistic and GPEH measurements specified in [20], whichhas been presented in Sections 4 and 5.
4. Performance management statistic
Performance management statistic has been obtained by RNCperformance management counters, pmSumUlRssi and pmSam-plesUlRssi that exist for every cell controlled by RNC:
� pmSumUlRssi is counter that records the value of received totalwideband power (RTWP) measured on cell and sent to the RNCvia nodeB application part (NBAP) common measurementreporting [21], over Iub interface which is specified in [22],where RTWP refers to UL RSSI measurement.
� pmSamplesUlRssi is a counter that records how often samplehas been recorded and it is actually captured every 10 s. Thatmeans that RNC is notified every 10 s of measured UL RSSI ateach cell on RNC.
Due to large amount of data, the obtained results have beenaggregated by RNC. Aggregated UL RSSI value over 15 min has beenstored on performance management statistic database, for everyWCDMA cell on RNC. The obtained result by performance manage-ment statistic, for one real WCDMA cell with typical data traffic,has been presented at Fig. 3.
As it can be concluded from Fig. 3, there are exactly 96 measure-ment results during monitoring period of one day, thus highest fre-quency for sirMax change could be every 15 min. From (1), self-optimizing uplink outer loop power control, based on performancemanagement statistic, would use 96 sirMax values during one day.
As it has been presented in this section, RNC is notified every10 s of measured UL RSSI at each cell on RNC, which is system limitand highest possible frequency for obtaining UL RSSI data. Theinformation of measured UL RSSI in 10 s resolution can be obtainedby GPEH measurements, which has been presented in next section.
5. GPEH measurements
The purpose of GPEH measurements has been providing real-time performance monitoring capabilities instead of aggregatingdelay introduced by performance management statistic. The col-lection and recording of GPEH measurements has been instru-mented by monitoring tasks which are accessed throughmanagement interfaces supporting the CORBA (Common ObjectRequest Broker Architecture) set of the performance and notifica-tion integration reference points. GPEHmeasurements for WCDMAcan monitor:
� Control functions.� Mobility functions.� Capacity functions.� Bearer functions.� Measurement reports.� Configuration management functions.� Inter-node communication.
Fig. 3. Measured UL RSSI on monitored cell obtained by pmSumUlRssi and pmSamplesUlRssi.
Fig. 4. Measured UL RSSI on monitored cell obtained by GPEH measurements.
1222 A.G. Markoc, G. Šišul / Int. J. Electron. Commun. (AEÜ) 70 (2016) 1219–1227
The main drawback of using GPEH compared to performancemanagement statistic is large amount of data that has been col-lected and stored on database. The result of measured UL RSSIobtained by GPEH measurements, for same real WCDMA cell fromprevious section, in same time period, has been presented at Fig. 4.
As it can be concluded from Fig. 4, there are exactly 8640 mea-surement results during monitoring period of one day, thus highestfrequency for sirMax change could be every 10 s. From (1), self-optimizing uplink outer loop power control, based on GPEH mea-surements, would use 8640 sirMax values during one day.
Furthermore, required uplink SIR target has been obtained byGPEHmeasurements for all data connections on same real WCDMA
cell in same time period. The obtained results have been presentedat Fig. 5.
Presented results for measured UL RSSI and required uplink SIRtarget have been used to investigate optimal implementation fordynamic change of sirMax, which has been presented in nextsection.
6. Optimal implementation for dynamic change of sirMax
As it can be concluded from previous sections live data havebeen used for evaluation purpose. Live data can be obtained byGPEH measurements on 10 s resolution or by performance
Fig. 5. Distribution of required uplink SIR target, on monitored cell, obtained by GPEH measurements.
Fig. 6. Distribution of required uplink SIR target and implemented fixed sirMax value.
A.G. Markoc, G. Šišul / Int. J. Electron. Commun. (AEÜ) 70 (2016) 1219–1227 1223
management statistic with highest resolution of 15 min (other res-olutions are hour, day, week and month). Analysis has been madeto determine would it be good enough to implement changes
based on UL RSSI from performance management statistic (whichis simpler), or there is actual need to use highest possible resolu-tion and GPEH measurements which is more complicated.
Fig. 7. Required SIR target compared to implemented fixed sirMax value of 12 dB.
Fig. 9. Required SIR target compared to variable sirMax value based on performancemanagement statistic.
1224 A.G. Markoc, G. Šišul / Int. J. Electron. Commun. (AEÜ) 70 (2016) 1219–1227
To determine optimal implementation of proposed algorithm ithas been compared to existing uplink outer loop power controlwith fixed sirMax implementation of 12 dB, which is optimizedvalue by network operator for normal cell load, presented at Fig. 6.
As it can be concluded from Fig. 6, there have been data connec-tions with required SIR target above implemented sirMax value.Thus, uplink throughput has been limited for those data connec-tions, which has been presented at Fig. 7.
As it can be concluded from Fig. 7, there have been 22% dataconnections with limited uplink throughput.
To determine optimal implementation for dynamic change ofsirMax, variable sirMax has been obtained from (1). Used Max-ULRSSI for self-optimizing uplink outer loop power control hasbeen determined from (2) and set to �88dBm, while C has been�2 dB. The result for variable sirMax, with ULRSSI from Fig. 3, hasbeen presented at Fig. 8.
Fig. 8. Variable sirMax based on UL RSSI fro
Self-optimizing uplink outer loop power control, based on ULRSSI from performance management statistic, has produced 96 sir-Max values. Compared to fixed sirMax value of 12 dB all 96 variablesirMax values have been above 12 dB. Thus number of data connec-tions with limited uplink throughput has changed, which has beenpresented at Fig. 9.
As it can be concluded from Fig. 9, number of data connectionswith limited uplink throughput has decreased from 22% to 8%. Onthe other hand, there has been no additional throughput limitationdue to that all 96 variable sirMax values have been above fixed sir-Max value. Thus it can be concluded that uplink overload, duringmonitored period, has not been detected by self-optimizing uplink
m performance management statistic.
Fig. 10. Variable sirMax based on UL RSSI from GPEH measurements.
Fig. 11. Required SIR target compared to variable sirMax value based on GPEHmeasurements.
Fig. 12. Uplink overload detection by variable sirMax implementation based onGPEH measurements.
A.G. Markoc, G. Šišul / Int. J. Electron. Commun. (AEÜ) 70 (2016) 1219–1227 1225
outer loop power control based on UL RSSI from performance man-agement statistic.
Same analysis has been performed for self-optimizing uplinkouter loop power control based on UL RSSI from GPEH measure-ments. Variable sirMax has been obtained from (1), withMaxULRSSIset to �88dBm, while C has been �2 dB and ULRSSI has beenobtained from Fig. 4. The result for variable sirMax, based on ULRSSI from GPEH measurements, has been presented at Fig. 10.
Self-optimizing uplink outer loop power control, based on ULRSSI from GPEH measurements, has produced 8640 different sir-Max values. Compared to fixed sirMax value of 12 dB, 8350 variable
sirMax values have been above, while 290 variable sirMax valueshave been below 12 dB. Thus number of data connections withlimited uplink throughput has changed, which has been presentedat Fig. 11.
As it can be concluded from Fig. 11, number of data connectionswith limited uplink throughput has decreased from 22% to 8%,which is same improvement as in case of self-optimizing uplinkouter loop power control based on UL RSSI from performance man-agement statistic. On the other hand, there has been additionalthroughput limitation due to 290 variable sirMax values that havebeen below fixed sirMax value of 12 dB. From obtained result it can
Table 1Achieved improvements by self-optimizing uplink outer loop power control based onsource of data for UL RSSI.
Source of data for UL RSSI
GPEH measurements Performance managementstatistic
Uplink capacityimprovement
Number of dataconnections with limiteduplink throughput hasdecreased from 22% to 8%
Number of dataconnections with limiteduplink throughput hasdecreased from 22% to 8%
Uplink overloaddetectionduring shorttime periods
Uplink overload has beendetected in 3.36% time ofmonitored period
Uplink overload has notbeen detected
1226 A.G. Markoc, G. Šišul / Int. J. Electron. Commun. (AEÜ) 70 (2016) 1219–1227
be concluded that uplink overload, during monitored period, hasbeen detected, which has been presented at Fig. 12.
As it can be concluded from Fig. 12, uplink overload have beendetected in 3.36% of time of monitored period which correspondedto 290 short bursts of uplink overload with duration of 10 s each.
6.1. Analysis of recorded results
Improvement by self-optimizing uplink outer loop power con-trol compared to legacy uplink outer loop power control, basedon source of data for UL RSSI measurement, has been presentedat Table 1.
As it can be concluded from Table 1, the main difference hasbeen in cases of uplink overload during short time periods, whichhas not been detected by performance management statistic. Thusself-optimizing uplink outer loop power control based on UL RSSIfrom performance management statistic could set to high valuefor sirMax, which could lead to uplink instability. On the otherhand, self-optimizing uplink outer loop power control based onUL RSSI from GPEHmeasurements has successfully detected uplinkoverload during short time periods. Thus sirMax has been adjustedto appropriate value and possible uplink instability has beenavoided. For rapid changes of sirMax value, RNC signaling and pro-cessor load must be taken into the consideration. By implementa-tion of self-optimizing outer loop power control, sirMax would becalculated from (1) and it would not be any more RNC parameter.New parameters on RNC would be MaxULRSSI and C which arefixed values in a same way as it was sirMax in legacy implementa-tion. In this case additional RNC signaling and processor load hasnot been expected. Thus highest possible frequency for sirMaxchange would be recommended, which is equal to every 10 s.
Self-optimizing uplink outer loop power control with changesof sirMax value for every available UL RSSI measurement onWCDMA cell would be adapted to dynamic conditions on uplinkair interface. Thus in cases of low uplink load, higher value of sir-Max would be used, allowing higher radiated power per mobilestation and higher uplink throughput. On the other hand, in casesof high uplink load, lower value of sirMax would be used, thus lim-iting radiated power per mobile station and uplink throughput. Thechanges of sirMax in cases of high uplink load would be fast enoughto adapt to short bursts of uplink overload.
7. Conclusion
Self-optimizing uplink outer loop power control has been intro-duced to combat problems with uplink air interface overload inWCDMA networks. Dynamic change of sirMax based on currentuplink load has been introduced by new algorithm. Thus maximum
allowed radiated power per mobile stations has been dynamicallyadapted to current uplink load. In this paper optimal implementa-tion for dynamic change of sirMax has been investigated. Perfor-mance management statistic and GPEH measurements have beenused to obtain UL RSSI and required SIR target on monitoredWCDMA cell. Based on obtained results it has been concluded thatimprovement in percentage of data connections that wouldachieve higher uplink throughputs is same for changing sirMaxevery 15 min or every 10 s. The main difference has been in factthat short periods of uplink overload have been detected andresolved by changing sirMax every 10 s, which has not beendetected by changing sirMax every 15 min. High setting of sirMaxduring short periods of uplink overload could lead to uplink insta-bility of WCDMA cell. Thus it has been proposed, for novel WCDMAuplink outer loop power control algorithm, to change sirMax every10 s.
References
[1] Song L, Mandayan NB, Gajic Z. Analysis of an up/down power control algorithmfor the CDMA reverse link under fading. IEEE J Sel Areas Commun 2001;19(2):277–86. http://dx.doi.org/10.1109/49.914505.
[2] Kaur Khangura S, Kaur K, Uppal RS. Power control algorithms in wirelesscommunication. Int J of Comput Appl 2010;1(12):82–7. http://dx.doi.org/10.5120/253-410.
[3] Goldsmith A. Wireless communication. Cambridge; 2005. ISBN: 0521837162,9780521837163.
[4] Nasreddine J, Nuaymi L, Lagrange X. Adaptive power control algorithm withstabilization zone for third generation mobile networks. J Ann DesTelecommun 2006;61(9):1193–211. http://dx.doi.org/10.1007/BF03219888.
[5] Gunnarsson F, Gustafsson F. Control theory aspects of power control in UMTS.Control Eng Pract 2003;11(10):1123–5. http://dx.doi.org/10.1016/S0967-0661(03)00062-5.
[6] Singla P, Saxena J. Power control analysis of Gaussian approximated WCDMAsystem under Rayleigh fading channel. In: Int Conf on Comput, Commun &Autom (ICCCA). p. 1256–63. http://dx.doi.org/10.1109/CCAA.2015.7148612.
[7] Dadhich A, Agrawal S. Comparative study of various power control algorithmsin WCDMA. Int J Digit Appl Contemp Res 2012;1(4):292–4. ISSN: 2319-4863.
[8] Al Mamun A, Islam S, Yesmin F, Akter M, Jahan SA. Novel adaptive step powercontrol algorithm for 3G WCDMA cellular system. In: 12th Int Conf on Computand Inf Technol (ICCIT ’09). p. 526–30. http://dx.doi.org/10.1109/ICCIT.2009.5407294.
[9] De Lima CHM, Larsson K, Okvist P. Outer-loop power control based on hybridARQ protocol for WCDMA enhanced uplink. In: Int Telecommun Symp. p.682–7. http://dx.doi.org/10.1109/ITS.2006.4433360.
[10] Wigren T. WCDMA uplink load estimation with generalized rake receivers. VehTechnol 2012;61(5):2394–400. http://dx.doi.org/10.1109/TVT.2012.2190949.
[11] Markoc AG, Sisul G. Self-optimizing uplink outer loop power control forWCDMA network. Radioengineering 2015;24(2):499–506. http://dx.doi.org/10.13164/re.2015.0499.
[12] Ranhema M. UMTS network planning, optimization, and inter-operation withGSM. Singapore, Singapore: Wiley; 2008. 047082302X, 9780470823026.
[13] Alami FT, Aknin N, El Moussaoui A. Capacity estimation of multi-servicecellular network. Int J Comput Sci Eng (IJCSE) 2011;3(3):1363–8. ISSN: 0975–3397.
[14] Wigren T. Recursive noise floor estimation in WCDMA. Veh Technol 2010;59(5):2615–20. http://dx.doi.org/10.1109/TVT.2010.2045139.
[15] Holma H, Toskala A. WCDMA for UMTS: HSPA evolution and LTE. Chichester,England: Wiley; 2010. ISBN: 1119991900, 9781119991908.
[16] rd Generation Partnership Project (3GPP), Technical Specification (TS) 25.922V7.1.0 (2007-03). Radio resource management strategies. http://www.3gpp.org/DynaReport/25922.htm.
[17] Pusnik T, Skok G, Ivandic I. High capacity event network planning – Poljud. In:Proc of the 56th Int Symp ELMAR. p. 63–7. http://dx.doi.org/10.1109/ELMAR.2014.6923316.
[18] Klozar L, Polak L, Kaller O, Prokopec J. Effect of co-existence interferences onQoS of HSPA/WCDMA mobile networks. In: 23rd Int Conf on Radioelektron. p.312–5. http://dx.doi.org/10.1109/RadioElek.6530937.
[19] Tolstrup M. Indoor radio planning: a practical guide for GSM, DCS, UMTS, HSPAand LTE. England: Wiley; 2011. ISBN: 1119973686, 9781119973683.
[20] Gustas P, Magnusson P, Oom J, Storm N. Real-time performance monitoringand optimization of cellular systems. Ericsson Rev 2002;1(1):4–13.
[21] rd Generation Partnership Project (3GPP), Technical Specification (TS) 25.433V12.2.0 (2014-09). UTRAN Iub Interface Node B Application Part (NBAP)Signaling, http://www.3gpp.org/DynaReport/25433.htm.
[22] rd Generation Partnership Project (3GPP), Technical Specification (TS) 25.430V12.0.0 (2014-09). UTRAN Iub Interface General Aspects and Principles. http://www.3gpp.org/DynaReport/25430.htm.
A.G. Markoc, G. Šišul / Int. J. Electron. Commun. (AEÜ) 70 (2016) 1219–1227 1227
Angelo Gary MARKOC was born in Euclid in 1982, USA.He received his B.Sc. degree in Electrical Engineeringfrom Faculty of Electrical Engineering and Computing in2005. He is a frequency and parameterization planningexpert in Radio Access Network Development Depart-ment at the Vipnet d.o.o. in Zagreb. His main researchinterests are in testing and implementation of newWCDMA and LTE functionalities with special focus increating and designing algorithms that automaticallyoptimize network performances based on inputs col-lected from live network. He published several papers innational and international conferences and journals.
Gordan ŠIŠUL received B.Sc., M. Sc. and Ph.D. in elec-trical engineering from the Faculty of Electrical Engi-neering, University of Zagreb, Croatia in 1996, 2000 and2004 respectively. He is currently employed as anAssociate Professor at the same Faculty. His academicinterests include wireless communications, signal pro-cessing applications in communications, modulationtechniques and radio propagation.
