Pattern synthesis of array antennas using multi-objective invasive weed optimization based on decomposition

The traditional pattern synthesis of array antennas is analyzed and solved as a multi-objective optimization problem, and a new algorithm called multi-objective invasive weed optimization based on decomposition(MOEA/D-IWO) is proposed by integrating the improved invasive weed optimization algorithm into the framework of the multi-objective evolutionary algorithm based on decomposition. The proposed algorithm completes the parallel calculations efficiently, through making good use of the powerful searching ability and robustness of invasive weeds. Compared with multi-objective differential evolution based on decomposition (MOEA/D-DE), synthesis results for a 20 element linear array show that the array obtained by the proposed algorithm has 1.582 2~2.115 1 dB sidelobe level reduction, 4.429 6 dB nulls reduction and 4.665 7 dB notches reduction. Through the experiments, MOEA/D-IWO shows better performance in computation accuracy, convergence speed and solution diversity.