Space-time processing with block shanno algorithm for smart antennas in ds-cdma systems
Dae-Hyun Paik, Richard M. Golden, Murat Torlak · 2004
Space-time blind adaptive processing algorithms are proposed for the uplink base station receiver in Direct Sequence Code Division Multiple Access (DS-CDMA) cellular system where its performance is interference-limited. The proposed algorithm, called Space-Time Block Shanno Constant Modulus Algorithm (STBSCMA), minimizes a CMA objective function in a bilinear function structure employing soft constraints for both spatial domain and temporal domain. STBSCMA attains superb performance on the nonlinear objective function for blind adaptive processing. The soft constraint for spatial domain tends to constrain the magnitude of the spatial filter tap weight vector, while the soft constraint for temporal domain tends to orient the temporal filter tap weight vector towards the user's signature sequence. These constraint terms serve two functions. First, they inform the algorithm regarding the desirability of particular solutions. Second, they address the local-identifiability problem in the sense that it can be shown that if the penalty parameters which indicate the influence of the soft constraints are sufficiently large, then critical points which have a projection upon the user's signature sequence are forced to become strict local minima of the modified objective function. It is shown that the local identifiable problem can not only be addressed (reducing the tendency of algorithm wandering) but also the modifications to the original objective function may be judiciously chosen to inform the algorithm which solutions are more desirable. It is shown that STBSCMA is able to converge rapidly and recovers fast from the abrupt change of channel with robustness in near-far environments of the DS-CDMA system. STBSCMA shows very rapid convergence with a relatively low complexity which is close to the level of narrowband processing rather than the level of wideband processing. STBSCMA successfully suppresses the ambient interferences in temporal domain and increases signal-to-interference-noise-ratio by separating the desired user from the interfering users in spatial domain. The STBSCMA is effectively applicable for the future generation mobile networks such as wireless internet systems and telematics systems where the environments are usually non stationary and varying rapidly.