Blind detection of dual-rate DS-CDMA signals over time-variant channels

To accommodate information streams with different data rates over a direct-sequence code-division multiple-access (DS-CDMA) network is required due to a large variety of information sources, such as voice, data, and images. This paper provides a novel adaptive blind estimating method for dual-rate DS-CDMA signals. The underlying idea is to estimate the channel and spatial signature by transforming them into one composite channel, which is estimated with using linear constrained minimum variance filtering and the method of the Lagrange multipliers. The simulation results show some improvements of the bit-error rate performance over the conventional detector using the subspace method and also confirm that the proposed method is near-far resistant for both cases with and without inter-chip interference.