Pilot-aided ML Schemes for Joint Beamforming and Channel Estimation in WCDMA Systems

This paper p roposes semi-blind channel e s t ima t ion and interference cancel la t ion schemes for the reception of pilot-aided signals i n WCDMA systems. It is shown that the pe r fo rmance of classical training-based schemes is severely degraded due t o the code-multiplexing (as opposed to t imemult iplexing) of traffic and pilot signals. The timedispers ivi ty of the channel des t roys the orthogonality be tween the trafiic signal and the pilot , a n d consequen t ly the traffic signal appears as a new interfer ing signal for the channel es t ima t ion /beamforming algorithm. In order to avoid this effect, w e propose techniques that exploi t b o t h the presence of the t r a in ing sequence and the structure of the traffic signal in a semi-blind fashion.