A novel interference rejection scheme for DS-CDMA using adaptive noise cancellation

This paper proposes a novel solution to the interference cancellation problem in mobile direct sequence code division multiple access (DS-CDMA) systems. Conventional adaptive interference cancellation techniques rely on a training sequence to update the taps of an adaptive filter or decision feedback equaliser (DFE). The strategy proposed in this paper replaces this conventional optimal filtering approach by one based on the principles of adaptive noise cancellation (ANC). As opposed to the conventional optimal filtering approach, the ANC approach exploits the cyclo-stationary properties of the multiple access interference, in order to model the interference generation process. Silent periods (c.f. voice activity factor) can be exploited to derive the interference model. This removes the need for a training sequence and associated system overheads. The scheme employs a conjugate matched filter (CMF) to generate an interference reference input to the adaptive noise canceller. A preliminary investigation of the performance of the proposed scheme is undertaken. The ANC scheme is shown to have a significantly better performance than the conventional receiver and the DFE in the multiple access interference limited environment. It is shown that the scheme can be extended to provide blind interference cancellation in the case of acute near-far conditions