Beamforming Optimization by Binary Genetic Algorithm

This paper presents a beamforming method based on a binary genetic algorithm. Widely known for their ability to increase the performance of antenna arrays, beamforming techniques are expected to play a major role in 5G systems. For reaching low peaks side lobe level (PSLL) in antenna design, it is necessary to optimize the amplitude weights of a linear antenna array. By optimizing the amplitude weight of the array's elements, a method for achieving a low side lobe level is explored. Utilized is a binary genetic algorithm with single point crossover and roulette wheel selection. The minimum SLL (minimize side lobe levels) for the radiation pattern is the cost function that is employed. The convergence of the optimization algorithm is demonstrated by simulations under Matlab environment and, its utility is shown in getting a desired antenna beam pattern, BGA converges before 50 generations in all the considered cases, We have seen that the level of the secondary lobes reaches nearly −30 dB for all the networks. We also note that the more the size of the antenna array increases, the more the number of iterations necessary for convergence is high.