Conformal Array Pattern Synthesis Using a Hybrid GA/ABC Algorithm

Abstract

A hybrid evolutionary algorithm is presented here to solve the problem of conformal array pattern synthesis. In order to overcome the disadvantage of standard genetic algorithm (GA) and artificial bee colony algorithm (ABC), a hybrid algorithm is introduced, which combines GA and ABC to take advantages of both methods. Crossover operator of GA is adopted to maintain the diversity of population. Multi-dimensional neighborhood search strategy of ABC is introduced to improve the local search efficiency for conformal array pattern synthesis. Finally, the hybrid GA/ABC algorithm is used to optimize the weight vector of the circular conformal array. Experimental results show that the proposed method can achieve the desired pattern very well, and has a better performance than standard GA and ABC. In this paper, a hybrid GA/ABC algorithm is introduced for pattern synthesis of conformal arrays to take the advantages of both algorithms. Crossover operator of GA is adopted to maintain the diversity of population. Multi-dimensional neighborhood search strategy of ABC is introduced to improve the local search efficiency for conformal array pattern synthesis. The hybrid algorithm is used to search the optimal weight vector of conformal array to achieve a low side lobe level pattern with the constraint of main lobe shaping. At last, circular conformal array is used as an example to verify the effectiveness of proposed method. Simulation results illustrate the superior performance compare to GA and ABC.