Using artificial bee colony algorithm with crossover for power allocation in cognitive MIMO-OFDM system

Artificial bee colony (ABC) algorithm builds on simulating the intelligent behavior of honey bees. It shows good performance in many applications. As standard ABC algorithm does not employ any crossover operator, the dispersal of good genetic information amongst the solutions is undermined. In this paper, the impact of crossover operators on the performance of ABC is studied. Eight crossover operators, representing all kinds of crossover operators, are used in this study. A trial and error method is used to detect the most proper crossover operator and crossover rate for incorporation into the ABC algorithm on mathematical functions as an initial attempt. The overall best configuration of ABC with crossover which has been identified is then applied to solve power allocation problem in cognitive multiple input and multiple output orthogonal frequency division multiplexing (MIMO-OFDM) cognitive system. Promising performances are obtained when compared with those from genetic algorithm, particle swarm optimization and differential evolution algorithm.