Stochastic analysis of an iterative semi-blind adaptive beamforming algorithm

We present an iterative semi-blind adaptive beamformer for time-division multiple-access (TDMA) cellular systems and analyze its performance using a stochastic model. The array inputs are iteratively processed using re-encoded data with blind adaptation and several stages that include beam-forming, demodulation, equalization, and decoding. In each stage, the initial beamformer weights computed by the TDMA training data are refined by a semi-blind technique based on the constant modulus algorithm (CMA). The receiver algorithms are verified using simulated and real TDMA data collected by an eight-element antenna array. The stochastic analysis of the iterative semi-blind beam-forming algorithm is based on a fourth-order moment, and demonstrates the improved performance of the receiver.