基于文化粒子群算法的广义OPM测向

在引入合成导向矢量和合成导向矩阵的基础上建立了一种更通用的阵列数据模型,提出了广义正交传播算子测向算法。该方法在不损失阵列孔径的前提下可有效估计相干信源和独立信源。为快速求解所提基于四阶累积量的广义OPM算法,在文化算法中使用和粒子群演进进化机制,提出了一种多维搜索的文化粒子群算法。Monte-Carlo仿真试验证明了所设计的测向算法可有效解决使用高阶累积量不能直接测相干信源的局限,所提算法在检测性能上与现有一些算法比较有较大的优势。

Generalized OPM Algorithm Based on Cultural Particle Swarm Optimization for Direction Finding

To resolve the disadvantages of direction finding algorithm in the coherent source and the computational complexity,a generalized data model is constructed based on generalized steering vectors and steering matrix.And a generalized orthogonal propagator method（GOPM） is proposed for direction finding.The proposed GOPM algorithm could estimate coherent sources and non-coherent sources without using more sensors.For fitting the cost function of GOPM,a cultural particle swarm optimization（CPSO） algorithm is proposed based on cultural algorithm and particle swarm optimization algorithm for optimal value searching.The CPSO algorithm can resolve real value optimization problem with high efficiency.Monte-Carlo simulations proves that the proposed algorithm not only can find direction of coherent sources in the algorithms based on fourth order cumulates,and also is better than the previous algorithms in performance.