Minimum BER power allocation for MIMO spatial multiplexing systems

Power allocation for multiple-input multiple-output (MIMO) spatial multiplexing systems is investigated. Minimization of bit error rate (MBER) is employed as the optimization criterion. MBER power allocation schemes for a variety of receiver structures, including zero-forcing (ZF), successive interference cancellation (SIC), and ordered SIC (OSIC), are proposed. It is shown that the newly proposed MBER power allocation schemes improve error rate performance. Simulation results indicate that SIC and OSIC with MBER power allocation outperform MBER precoding for ZF equalization as well as minimum mean squared-error (MMSE) preceding/decoding. Performance under noisy channels and power feedback is analyzed. A modified algorithm that mitigates error propagation in interference cancellation is developed. Compared to existing precoding methods, the proposed schemes significantly reduce both processing complexity and feedback overhead, and improve error rate performance.