Optimum Signature Sequence Sets for Asynchronous CDMA Systems

In this paper we characterize the user capacity i e the maximum number of supportable users at a common SIR target level for a xed processing gain of a single cell symbol asynchronous CDMA system Based on the delay pro le of the users we identify a class of optimum signature sequences that achieve a lower bound on the total squared asynchronous correlation TSAC among the users When the users signatures achieve this lower bound the user capacity of a single cell asynchronous CDMA system becomes the same as that of a single cell synchronous CDMA system that is there is no loss in user capacity due to asynchrony Further when the optimum signature sequences are used the users received powers are all equal and the M shot MMSE receiver lters turn out to be scaled matched lters That is the maximum user capacity is achieved by observing only one symbol interval of the received signal and using single user matched lters in that interval It is a signi cant open question whether these optimal sequence sets exist for all delay pro les However we present iterative and distributed signature adaptation algorithms which when executed sequentially by all of the users appear to converge to these optimum signature sequences