Evaluation of New NLMS and RLS Chip Equalizers using Tentative Decision Data for HSDPA Systems

Abstract

Equalizers are currently being studied to improve the performance of wideband code division multiple access (WCDMA) downlinks. Especially for high data rate wireless systems (e.g. high-speed downlink packet access (HSDPA)), reduction of multiple access interference (MAI) and inter symbol interference (ISI) are key issues. We propose a new chip equalizer using tentative decision data and show that it has a high throughput performance for HSDPA systems. The normalized least mean square (NLMS) algorithm or the recursive least square (RLS) algorithm is used as the adaptive algorithm for the equalizer because these equalizers are well known to have lower complexity than the maximum mean square error (MMSE) equalizer. In the proposed method, high-speed physical downlink shared channel (HS-PDSCH) data is tentatively determined and used as a training signal to add to the common pilot channel (CPICH). In this paper, we show that the proposed equalizer has a good throughput performance compared with the conventional equalizer, and that the proposed RLS equalizer has a throughput performance that is approximately 73% of the ideal MMSE.