Design of Dual Pattern Concentric Ring Array Antenna Using Differential Evolution Algorithm with Novel Evolutionary Operators

Concentric ring antenna arrays with the ability to produce dual pattern have many applications in communications and radar engineering. In this paper, we represent a new method for design of an optimized reconflgurable concentric ring array with dual pattern of desired speciflcations. Here, our goal is to flnd a suitable common element excitation amplitude distribution and two difierent element excitation phase distributions for two desired radiation patterns. For this purpose, we have proposed a novel objective function which is completely difierent from the traditional objective functions usually used in antenna design problems. For the optimization procedure, we have developed a modifled Difierential Evolution (DE) algorithm, denoted as DE rBM 2SX, which employs new kinds of crossover and mutation operators to overcome some drawbacks of the classical DE on single-objective fltness landscapes. We consider three types of dual pattern | pencil beam+pencil beam, pencil beam+∞at-top beam, ∞at-top beam+∞at-top beam. The simulation results obtained by applying our proposed method clearly indicate that our method is very convenient to obtain radiation patterns of desired speciflcations. We compare results of the modifled DE algorithm with those of another state-of-the-art improved variant of DE, called JADE and a state-of- the-art variant of the Particle Swarm Optimization (PSO) algorithm called Comprehensive Learning Particle Swarm Optimizer (CLPSO). Such comparisons re∞ect that the proposed algorithm is more e-cient than JADE or CLPSO in flnding optimum conflguration of the dual pattern concentric ring array antenna.