A semiblind adaptive antenna for WCDMA using a least-squares constant-modulus formulation

This paper presents a semiblind adaptive antenna for the downlink of wideband code-division multiple-access (WCDMA) systems. The weight vector of the antenna array is computed using the coded-multiplexed pilot channel as well as the estimated traffic channels in an extended training sequence. This approach suppresses the self-interference from physical channels in the same base station, leads to more accurate cochannel signal separation, and thus improves the performance compared to a conventional pilot-only trained beamformer. Moreover, a constant modulus constraint with a least-squares formulation is combined with the semiblind approach to refine the antenna weights. The performance of the proposed semiblind antenna system is evaluated via computer simulations.