Genetic algorithm implementation of multi-user detection in SDMA-OFDM systems

Number of supported users in the orthogonal frequency division multiplexing (OFDM) systems can be increased considerably using powerful multi-user detector (MUD) combined with space division multiple access (SDMA) techniques. This paper presents the results of implementing MUD in SDMA-OFDM systems based on an advanced genetic-algorithm (GA) optimization tool. The hardware implementation is performed using Field Programmable Gate array (FPGA) devices which allow the real time performance of the proposed tool. Results show that the GA scheme enhances the performance and provides BER near to that attained using maximum likelihood (ML) detector at considerably lower computation complexity. Investigation of the GA population size is presented and FPGA implementation is described based on the shared memory approach.