Genetic Algorithm-assisted Maximum-Likelihood Multi-user Detection for Multi-Rate MC-CDMA Systems

In this letter , the use of Genetic Algorithms (GAs) has been investigated in order to implement a computationally tractable Multi-User Detection (MUD) for synchronous multi-rate Multi-Carrier-Code-Division-Multiple-Access (MR MC-CDMA) systems. It is known that MC-CDMA performances can be seriously limited by multi-user interference. This is particularly true in the case of multi-rate systems where some users could exploit a reduced diversity against multipath channel distortions. The use of Genetic Algorithms (GAs) is therefore proposed as a sub-optimal solution to the MUD problem in a multi-rate Variable-Spreading-Length (VSL) MC-CDMA downlink transmission. A set of selected numerical experiments have been reported and discussed in order to assess the capability of the proposed GA-based approach. The reported results, obtained with an affordable computational burden, have shown a near-optimum behavior of the proposed GA-based MUD.