BER and MSE performance evaluation of adaptive interference suppression schemes in asynchronous CDMA systems

Includes bibliographical references (pages [82]-85)Multiple access (MA) interference leads to significant performance degradation in code division multiple access (CDMA) system. The MA interference is most severe when users have different power levels and/or when they transmit their signals asynchronously. The conventional matched-filter scheme cannot effectively suppress this kind of interference. Much effort has been made to develop a scheme that can suppress the severe MA interference. Among all different interference suppressing schemes, two minimum-mean-square-error (MMSE) based schemes, namely, N-tap scheme and D-tap cyclically shifted filter bank (CSFB) scheme, are studied. Both schemes require only modest computations and are amenable to adaptive implementations that do not require knowledge of the interference parameters such as relative strength and spreading sequences of the users. These two MA interference suppression schemes are implemented using several adaptive algorithms, namely, least mean square, gradient lattice and predictive least mean square. Their performance of suppressing the MA interference is evaluated using the bit-error-rate (BER) criterion as well as minimum-mean-square error (MMSE) criterion. Computer simulations are carried out to obtain the MSE and BER under various interference conditions. The results show that the N-tap scheme and D-tap CSFB scheme yield a much better performance than that of the matched-filter scheme.M.S. (Master of Science