A comparative study of heuristic algorithms: GA and UMDA in spatially multiplexed communication systems

作者：Bashir Sajid^*; Naeem Muhammad; Shah Syed Ismail

来源：Engineering Applications of Artificial Intelligence, 2010, 23(1): 95-101.

DOI：10.1016/j.engappai.2009.08.005

• Computational complexity
• Detection
• Genetic algorithm
• Multiple input multiple output
• Univariate marginal distribution algorithm
• Vertical bell labs space time architecture

A performance comparison of genetic algorithm (GA) and the univariate marginal distribution algorithm (UMDA) as decoders in multiple input multiple output (MIMO) communication system is presented in this paper. While the optimal maximum likelihood (ML) decoder using an exhaustive search method is prohibitively complex, simulation results show that the CA and UMDA optimized MIMO detection algorithms result in near optimal bit error rate (BER) performance with significantly reduced computational complexity. The results also suggest that the heuristic based MIMO detection outperforms the vertical bell labs layered space time (VBLAST) detector without severely increasing the detection complexity. The performance of UMDA is found to be superior to that of CA in terms of computational complexity and the BER performance.

• 出版日期2010-2