Pattern Synthesis Of Conformal Array Based On Intelligent Optimization Algorithms

Conformal array antenna is an antenna that can conform to a prescribed shape. Conformal arrays not only have a good aerodynamic performance, but also can save the carrier space and simplify the installation. So the conformal array antenna is widely applied in the fields of radar, communication and navigation etc. Pattern synthesis of antenna array is one of the conformal antenna techniques. So the research of antenna array pattern synthesis is of great significance. In this paper, the pattern synthesis problem of conformal arrays is studied with low side lobe under the main lobe constraint, and the intelligent optimization algorithm is applied to this problem. Two kinds of conformal array pattern synthesis algorithm are proposed based on artificial colony algorithm. The main contents are as follows:First of all, a basic model of the conformal array is established, and a far-field pattern function is given. Then the basic characteristics of uniform linear array and uniform circular array are analyzed. Several common methods of array pattern synthesis are introduced in this paper and the simulation experiments are carried on.Secondly, a low side lobe pattern synthesis algorithm is proposed based on the improved artificial bee colony algorithm. The problem is formulated as an optimization problem to minimize the distance from the obtained pattern to the desired one. Improved multi-dimensional neighborhood search strategy is introduced to improve the local search efficiency in the weight vector space, and a penalty function is adopted to suppress the side lobe level. With the optimal weight vector, the beam pattern can finally approach the desired one. Experimental results show that the improved algorithm has better convergence speed and stronger ability to suppress side lobe than the original algorithm, standard genetic algorithm and particle swarm algorithm.Finally, In order to overcome the disadvantage of standard artificial bee colony algorithm, a hybrid algorithm is introduced. Crossover operator of GA is adopted to enhance the global search ability and improve the convergence speed. The shape of main lobe and side lobe level is controlled by setting the upper and lower limits. Experimental results show that the proposed method can achieve the desired pattern very well and has a superior performance.