Abstract-In this paper, we propose an optimization method based on real-coded genetic algorithm (GA) with elitist strategy for thinning a large linear array of uniformly excited isotropic antennas to yield the maximum relative sidelobe level (SLL) equal to or below a fixed level. The percentage of thinning is always kept equal to or above a fixed value. Two examples have been proposed and solved with different objectives and with different value of percentage of thinning that will produce nearly the same sidelobe level. Directivities of the thinned arrays are found out and simulation results of different problems are also compared with published results to illustrate the effectiveness of the proposed method.

Abstract-In this paper, we present a comparison study between phase-only and amplitude-phase synthesis of symmetrical dual-pattern linear antenna arrays using floating-point or real-valued genetic algorithms (GA). Examples include a sum pattern and a sector beam pattern.In the former, phase is only optimized with predetermined Gaussian amplitude distribution of fixed dynamic range ratio (|a max /a min |) and in the latter, both are optimized with less dynamic range ratio than the former and yet share a common amplitude distribution.

In this paper, the authors present an optimization method based on modified Particle Swarm Optimization (PSO) algorithm for thinning large multiple concentric circular ring arrays of uniformly excited isotropic antennas that will generate a pencil beam in the vertical plane with minimum relative side lobe level (SLL). Two different cases have been studied, one with fixed uniform inter-element spacing and another with optimum uniform inter-element spacing. In both the cases, the number of switched off elements is made equal to 220 or more. The half-power beam width of the synthesized pattern is attempted to make equal to that of a fully populated array with uniform spacing of 0.5λ. Simulation results of the proposed thinned arrays are compared with a fully populated array to illustrate the effectiveness of our proposed method.

Abstract-This paper describes the application of two recently developed metaheuristic algorithms known as fire fly algorithm (FFA) and artificial bees colony (ABC) optimization for the design of linear array of isotropic sources. We present two examples: one for broad side arrays and the other for steerable linear arrays. Three instances are presented under each category consisting of different numbers of array elements and array pattern directions. The main objective of the work is to compute the radiation pattern with minimum side lobe level (SLL) for specified half power beam width (HPBW) and first null beam width (FNBW). HPBW and FNBW of a uniformly excited antenna array with similar size and main beam directions are chosen as the beam width constraints in each case. Algorithms are applied to determine the non-uniform excitation applied to each element. The effectiveness of the proposed algorithms for optimization of antenna problems is examined by all six sets of antenna configurations. Simulation results obtained in each case using both the algorithms are compared in a statistically significant way. Obtained results using fire fly algorithm shows better performances than that of artificial bees colony optimization technique provided that the same number of function evaluations has been considered for both the algorithms.

Abstract-Scanning a planar array in the x-z plane directs the beam peak to any direction off the broadside along the same plane. Reduction of sidelobe level in concentric ring array of isotropic antennas scanned in the x-z plane result in a wide first null beamwidth (FNBW). In this paper, the authors propose pattern synthesis methods to reduce the sidelobe levels with fixed FNBW by making the scanned array thinned based on two different global optimization algorithms, namely Gravitational Search Algorithm (GSA) and modified Particle Swarm Optimization (PSO) algorithm. The thinning percentage of the array is kept more than 45 percent and the first null beamwidth (FNBW) is kept equal to or less than that of a fully populated, uniformly excited and 0.5λ spaced concentric circular ring array of same scanning angle and same number of elements and rings.