Improved Signal Detection Approach using Genetic Algorithm for Overloaded MIMO Systems

Abstract

The genetic algorithm (GA) is a well studied technique, which can obtain efficiently a near-optimal solution with much lower computational complexity compared to the maximum likelihood detection (MLD) approach. In overloaded MIMO systems, since the number of near equal-power co-channel interferers is higher than the number of receive antennas, the interference cannot be completely removed by spatial filtering. In this paper, we propose a new GA based detection approach for overloaded MIMO systems, whereby proportional scaling and Gaussian mutation is employed with a hybrid function to achieve a better performance than rank scaling and uniform mutation in the previous work. Simulation results show that our proposed approach not only significantly outperforms the zero-forcing (ZF) and minimum mean square error (MMSE) approaches and has much lower error floor, but also provides a performance which is close to the MLD with huge symbolic and numerical complexities reduction.