天線位置擾動下使用混合式粒子群最佳化結合引力搜尋法之陣列校正

Array Calibration Using Hybrid PSOGSAApproach under Array Position Errors

In this paper, we propose an array calibration method that uses the hybrid population-based algorithm (PSOGSA) with the combination of particle swarm optimization (PSO) and gravitational search algorithm (GSA). We consider the problem of estimating the direction-of-arrival (DOA) based on maximum likelihood (ML) criteria for code-division multiple access (CDMA) signals under sensor position errors. ML is a complex nonlinear, multimodal function that features high dimensional problem spaces. However, when the random array position errors occur, the performance degrades rapidly. In this paper, the main aim is to integrate the exploitation ability of PSO with the exploration ability of GSA and synthesize the strengths of both algorithms. The proposed technique offers a much faster convergence compared with either PSO or GSA alone. The proposed method has no requirement for calibrating sources, and the sensor position errors as well as the DOAs of the incident signals can be estimated simultaneously. Computer simulations are conducted to demonstrate the validity and feasibility of the proposed method.