Skip to Main Content
In the realization of adaptive beamforming algorithms, the least-mean-squares (LMS) algorithm had been the most popular scheme used for smart antenna systems. This paper proposes an alternate scheme in the form of the normalized LMS (NLMS) algorithm with active tap detection for WCDMA systems. By taking advantage of spatial filtering, the proposed scheme promises to reduce the bandwidth required for transmitting data by improving convergence rate. The performance of the frequency domain NLMS algorithm in the presence of multipath effects and multiple users is analyzed using simulations. This analysis is compared to that of LMS algorithm and suggests improvement in the convergence rate and number of active taps used, which leads to better system efficiency.
Date of Conference: 14-17 May 2007