Hybrid Differential Evolution and Enhanced Particle Swarm Optimization Technique for Design of Phased Antenna Arrays

Abstract

Abstract: In the last decades of the twentieth century, the use of evolutionary algorithms has been widely spread in the field of phased antenna array design. These algorithms, such as genetic algorithms (GA), simulated annealing (SA), particle-swarm optimization (PSO), and differential evolution (DE), have been successfully used in pattern synthesis applications. This thesis introduces methods for designing reconfigurable phased arrays using a novel hybrid differential evolution (DE) and enhanced particle swarm optimization (EPSO) technique. The proposed technique combines DE and an enhanced version of the standard PSO with an improved mechanism that updates velocity and global best solutions. In the hybrid algorithm, DE and EPSO are executed in parallel with frequent information sharing to enhance the newly generated population To demonstrate the effectiveness of the proposed algorithm over each separate algorithm, many examples were simulated using MATLAB. The first application was multiple null steering for linear and circular antenna arrays using DE, EPSO, and the newly proposed hybrid DE/EPSO. Nulls are controlled using a position control method by perturbing the position values of the selected arrays' elements, which affects the formation of the specified null locations in the radiation pattern. Also, the numbers of selected elements are optimized to have a minimum number of mobilized elements In the second application, the hybrid algorithm is used in designing different configurations of concentric circular arrays (CCA) and concentric hexagonal arrays (CHA) with a low sidelobe level (SLL) in the azimuth plane of the radiation pattern. Comparisons of different configurations having the same number of elements: two-ring, three-ring, and four-ring are presented. The outer rings of these configurations are rotated with optimized angles to minimize SLL. Both amplitude and phase (complex excitations) are set or optimized in an adequate way using the hybrid technique to achieve the design of the configuration with minimum SLL. Also, the effect of adding a central element to these configurations is studied. Amplitude dynamic range ratio and directivity are also studied for each configuration In the third application, the hybrid algorithm is used in designing time-modulated linear and planar arrays. The amplitude excitations of the time-modulated array only require a low dynamic range ratio or even a uniform distribution, while designing low/ultra-low sidelobe patterns is much easier for hardware implementation. The inherent drawback of time-modulated arrays is that there are many sideband signals spaced at multiples of the modulation frequency, which implies that part of the radiated or received power is shifted to the sidebands. The hybrid algorithm is used to optimize the switch-on times of the array elements to minimize the SLL and sideband level (SBL) that accompany the radiation pattern of the time-modulated antenna arrays. All results obtained using the hybrid methods are compared with the results of individual DE and EPSO methods. Simulation results demonstrate the refined global search capability and effectiveness of the proposed algorithm over each separate algorithm