Abstract
In this paper, we present three new methods to suppress narrowband interference in spread-spectrum code-division multiple-access (CDMA) systems, including a basic scheme and two performance-enhanced versions. The basic scheme is an adaptive nonlinear predictor that consists of an (N+1)-level quantizer, an adaptive linear filter, and two adders, where N is the number of users in the CDMA system. The first enhanced scheme is an adaptive nonlinear interpolator, and the second one is an adaptive nonlinear predictor with offset outputs; both involve a little bit more complexity than the basic scheme, and have their own distinct features in performance. Computer simulation results show that all the proposed approaches have much better performance than the conventional linear filtering approach. As compared to the approximate conditional mean (ACM) nonlinear filter, they also achieve similar performance when the SNR is large enough, but involve much less computational/hardware complexity.