Multi-user Detection in Ds-cdma Systems Using a Genetic Algorithm with Redundancy Saving Strategy

This study demonstrates that the configuration of a fixed generation number in the conventional genetic algorithm–based multi-user detector (GA-MUD) induces considerable redundant iterations and thus results in a computational inefficiency. To overcome this difficulty, this research suggests a redundancy saving strategy for the GAMUD (RSGA-MUD) based on the cost statistics of the GA solutions in a synchronous direct-sequence code division multiple access (DS-CDMA) system. Computer simulations for the synchronous DS-CDMA systems in perfect power control, near-far, and increased population scenarios are conducted to examine the performance of the proposed RSGA-MUD. Experimental results show that the RSGA-MUD can significantly reduce the computational burden of the conventional GA-MUD with only a slight performance degradation in perfect-power control and near-far scenarios. Furthermore, in the increased population scenario, the bit error rate (BER) performance is comparable to that of the optimum multi-user detector (OMD) with a computational complexity less than that of the conventional GA-MUD. Moreover, this study provides a novel mechanism that allows for a trade-off between computational reduction and acceptable degradation of GA-MUD performance by simply adjusting the cost threshold. Notably, the proposed redundancy saving strategy can be applied directly to various GA-based MUDs in order to increase their computational efficiency since it does not alter the essential structure of GA.