基于混合遗传算法的唯相位直接数据域算法

提出一种基于混合遗传算法的唯相位直接数据域最小二乘算法.通过采用标准遗传算法与Neider-Mead单纯形法相结合的混合遗传算法,提高了优化效率和运算速度.首先根据标准直接数据域算法推导得出目标函数,继而将目标函数作为适应度函数,将所有自适应权值的未知相位作为决策变量,通过混合遗传算法进行非线性优化,从而求得各个自适应权值的优化解.作为一种唯相位自适应算法,它在硬件实现上比传统算法更具简单性.同时,它只对单快拍数据进行处理,避免了样本协方差矩阵的构造以及矩阵求逆运算,更适合于实时处理.仿真结果表明,算法具有良好的信号恢复和干扰置零性能,比基于非线性其轭梯度法的唯相位直接数据域算法性能更优.

A Phase-Only Direct Data Domain Algorithm Based on Hybrid Genetic Algorithms

A phase-only D3LS(Direct Data Domain Least Squares) algorithm based on hybrid GA(Genetic Algorithms) is presented.The optimization efficiency and operational speed is improved via the hybrid GA composed of the standard GA and the Nelder-Mead simplex algorithm.The objective function is derived via the standard D3LS algorithm.With the objective function as the fitness function and the unknown phases of all adaptive weights as the decision variables,the nonlinear optimization is performed via the hybrid GA to obtain the optimized solution of adaptive weights.As a phase-only adaptive algorithm,the proposed algorithm has a better simplicity on hardware implementation than conventional algorithms.Moreover,it processes only a single snapshot data with no use for the formation of the sample covariance matrix and matrix inversion operation.Therefore,it may be so effective in real-time processing.Simulation results show the proposed algorithm has a good signal recovery and interferences nulling performance,which is superior to the phase-only D3LS algorithm based on the CG(Conjugate Gradient) method.