Optimizing time and space MIMO antenna system for frequency selective fading channels

Smart or adaptive antennas promise to provide significant increases in system capacity and performance in wireless communication systems. In this paper, we investigate the use of adaptive antennas at the base and mobile stations, operating jointly, to maximize the average signal-to-interference and noise ratio (SINR) of each packet in the system for frequency selective channels with prior knowledge of the channel at the transmitter. Our approach is based on deriving an analytic formula for the average packet SINR and using the Lagrange multiplier method to determine an optimum. We derive necessary conditions for an optimum solution and propose an analytical expression for the optimum. Our analytical expression is not guaranteed to be the global optimum but it does satisfy the derived necessary conditions and, in addition for frequency flat channels, our results reduce to expressions for optimal weights previously published, To demonstrate the potential of the proposed system, we provide Monte Carlo simulation results of the system bit-error rates and make comparisons with other adaptive antenna systems. These show that significant improvements in performance are possible in a wireless communications context.