TY  - JOUR
T1  - Performance of Cellular Neural Network Based Channel Equalizers
AU  - Özmen, Atilla
AU  - Tander, Baran
AU  - Şenol, Habib
PY  - 2020
DA  - January
DO  - 10.17694/bajece.519464
JF  - Balkan Journal of Electrical and Computer Engineering
PB  - MUSA YILMAZ
WT  - DergiPark
SN  - 2147-284X
SP  - 1
EP  - 6
VL  - 8
IS  - 1
LA  - en
AB  - Abstract—In this paper, a popular dynamic neural networkstructure called Cellular Neural Network (CNN) is employedas a channel equalizer in digital communications. It is shownthat, this nonlinear system is capable of suppressing the effect ofintersymbol interference (ISI) and the noise at the channel. Thearchitecture is a small-scaled, simple neural network containingonly 25 neurons (cells) with a neighborhood of r = 2 , thusincluding only 51 weight coefficients. Furthermore, a specialtechnique called repetitive codes in equalization process is alsoapplied to the mentioned CNN based system to show that thetwo-dimensional structure of CNN is capable of processing suchsignals, where performance improvement is observed. Simulationsare carried out to compare the proposed structures withminimum mean square error (MMSE) and multilayer perceptron(MLP) based equalizers.
KW  - Cellular Neural Networks
KW  - channel equalization
KW  - MLP equalizer
KW  - MMSE equalizer
CR  - C. Berrou, A. Glavieux, P. Thitimajshima, “”Near Shannon limit errorcorrectingcoding and decoding: Turbo-codes,” in Proceeding of EEEInternational Conference on Communications, Geneva, Switzerland,Nov. 30 - Dec. 4 1993, pp. 1064–1070.
CR  - M.K. Lee, K. Yang, “Scheduling for an adaptive number of iterationsin turbo equalizers combined with LDPC decoders,” IEEE T Commun,vol. 58, no. 10, pp. 2759–2764, 2010.
CR  - M. Tuchler, R. Koetter, A. Singer, “Turbo equalization: Principles andnew results,” IEEE T Commun, vol. 50, no. 5, pp. 754–767, 2002.
CR  - S. Talakoub S., L. Sabeti, B. Shahrrava, M. Ahmadi, “An improved maxlog-MAP algorithm for turbo decoding and turbo equalization,” IEEE TInstrum Meas, vol. 56, no. 3, pp. 1058–1063, 2007.
CR  - P. J.G., Digital Communications 4th ed. McGraw Hill, 2001.
CR  - J. Lee, C. Beach, N. Tepedelenlioglu, “A practical radial basis functionequalizer,” IEEE T Neural Networ, vol. 10, no. 2, pp. 450–455, 1999.
CR  - L. Biao, BL. Evans , “Channel equalization by feed forward neuralnetworks,” in In: Proceeding of IEEE International Symposium onCircuits and Systems, Orlando, FL, USA, May 30 - June 2 1999, pp.587–590.
CR  - R. Perfetti, “CNN for fast adaptive equalization,” J Circ Theor, vol. 21,no. 2, pp. 165–175, March 1993.
CR  - A. Ozmen, B. Tander, “Channel equalization with cellular neural networks,” in In: IEEE Mediterranean Electrotechnical Conference,Valletta, Malta, April 2010, pp. 1597–1599.
CR  - L.O. Chua, L. Yang, “A Cellular neural networks: Theory,” IEEE TransCAS, vol. 35, no. 10, pp. 1257–1272, 1988.
CR  - G. Costantini, D. Casali, M. Carota, “Detection of moving objects in 2-Dimages based on a CNN algorithm and density based spatial clustering,”WSEAS Trans CAS, vol. 4, no. 5, pp. 440–447, 2005.
CR  - K. S. M., Fundamentals of Statistical Signal Processing: EstimationTheory. Prentice Hall, 1993.
CR  - T. Tozek, T. Roska, L.O. Chua, “ Genetic algorithm for CNN templatelearning,” IEEE Trans. CAS I: Fund. Theo. and Apps, vol. 40, no. 6, pp.392–402, 1993.
UR  - https://doi.org/10.17694/bajece.519464
L1  - https://dergipark.org.tr/en/download/article-file/970148
ER  -