An Hlaing Minn creator Makelainen, Kimmo Chairperson Premanandana Rajatheva Examination committee Sharma, A.B. Examination Committee Government of Finland Scholarship donor text th Bangkok Asian Institute of Technology 1997 monographic eng

print

70 p.: ill. In performance evaluation of cellular mobile communication systems, simulation is a vital tool and usually conventional Monte Carlo simulation method is used. The simulation results are reliable but the simulation run time may be prohibitively long for very low BER estimation. Although BER of cellular mobile systems is not very low, faster simulation techniques are much desirable in system design because of its high complexity. Importance Sampling (IS) is the most popular fast simulation method but its applicability is quite limited due to its system specific nature. One main problem in IS is system memory or dimensionality and it has been successfully applied to the systems with high dimensionality such as Viterbi decoder and adaptive equalizer. In this study we try to extend the applicability ofIS to a system containing both adaptive equalizer and Viterbi decoder. One IS method, Stochastic Gradient Descent (SGD) algorithm, is applied to IS-54 TDMA cellular mobile communication system with Viterbi equalizer and Viterbi decoder. The simulation results show that SGD can be applied to the BER estimation of Viterbi equalizer output but does not perform well in application to the system containing cascaded Viterbi algorithms. A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering. Thesis (M.Eng.) - Asian Institute of Technology, 1997 Cellular radio Asian Institute of Technology. http://203.159.5.9/ait-thesis/detail.php?q=B14103 http://203.159.5.9/ait-thesis/detail.php?q=B14103 20260818144300.0