Performance evaluation of LMS-based adaptive suppression schemes in asynchronous CDMA

The objective of this paper is to evaluate the bit-error-rate performance of various adaptive structures in combating multi-user interference in an asynchronous code division multiple access (CDMA) system. In particular, two adaptive schemes, known as the N-tap filter and D-tap cyclical shifted bank filter, are considered. The least mean square (LMS) and predictive LMS (PLMS) algorithms are employed for the adaptation of tap weights. An analytical expression is developed for the numerical evaluation of the bit error probability. In addition, the bit error rate performances of the adaptive schemes are compared with those of the matched filter receiver. Results attest to the pratical usefulness of the LMS-based adaptive suppression schemes in combating multi-user interference in an asynchronous CDMA system.