A near maximum likelihood performance modified firefly algorithm for large MIMO detection

To meet the ever-growing demand for high data rates, employing a large number of antennas at both the transmitter and receiver is a necessity for future advanced wireless systems. Multiple-input multiple-output (MIMO) systems, which are equipped with multiple antennas, provide high data rates with high spectral efficiency. However, the design of an efficient, robust and non-erroneous detection algorithm is a huge challenge in large MIMO systems. In this paper, a stochastic bio-inspired meta-heuristic algorithm is proposed for large MIMO detection. The proposed algorithm is motivated by the bioluminescence of fireflies and uses a probabilistic metric to update solutions in the search space. Robustness of the proposed algorithm is verified under channel estimation errors at the receiver. Simulation results reveal that the proposed algorithm outperforms unordered congestion control ant colony optimization, congestion control ant colony optimization, standard particle swarm optimization, binary particle swarm optimization, memetic particle swarm optimization, firefly algorithm, firefly algorithm with neighborhood attraction, minimum mean square error and successive interference cancellation based MIMO detection techniques in terms of bit error rate (BER) performance. The proposed algorithm achieves near maximum likelihood BER performance with lower computational complexity. This makes the proposed algorithm an appropriate candidate for reliable detection in future large MIMO systems.