Sphere Decoding Assisted GA-based Multiuser Detection for Synchronous LAS Multicode System

Multicode system is designed to get more spreading sequences and get higher multiplexing capacity in large-area-synchronized code division multiple access (LAS-CDMA) system. The computational complexity is too high for the optimum detection algorithm using maximum-likelihood (ML) rule and the genetic algorithm (CA) method is often applied to get the suboptimum solution since it is able to greatly decrease the detection complexity. Another sphere decoding (SD) algorithm can also independently reach a near-optimum solution due to a constrained searching operation. A modified detection algorithm based on genetic algorithm method is proposed, where sphere decoding is an additional embedded operation for each CA iteration. The simulation results shows that a SD-assisted CA detector can obtain the same performance as that of the classic CA detector with fewer iterations.