Genetic algorithm assisted joint multiuser symbol detection andfading channel estimation for synchronous CDMA systems

摘要

A novel multiuser code division multiple access (CDMA) receivernbased on genetic algorithms is considered, which jointly estimates thentransmitted symbols and fading channel coefficients of all the users.nUsing exhaustive search, the maximum likelihood (ML) receiver innsynchronous CDMA systems has a computational complexity that isnexponentially increasing with the number of users and, hence, is not anviable detection solution. Genetic algorithms (GAs) are well known forntheir robustness in solving complex optimization problems. Based on thenML rule, GAs are developed in order to jointly estimate the users'nchannel impulse response coefficients as well as the differentiallynencoded transmitted bit sequences on the basis of the statisticsnprovided by a bank of matched filters at the receiver. Using computernsimulations, we showed that the proposed receiver can achieve annear-optimum bit-error-rate (BER) performance upon assuming perfectnchannel estimation at a significantly lower computational complexitynthan that required by the ML optimum multiuser detector. Furthermore,nchannel estimation can be performed jointly with symbol detectionnwithout incurring any additional computational complexity and withoutnrequiring training symbols. Hence, our proposed joint channel estimatornand symbol detector is capable of offering a higher throughput and anshorter detection delay than that of explicitly trained CDMA multiuserndetectors