三维参数联合估计的免疫记忆量子克隆算法

Immune Memory Based Quantum Clone Algorithm for Joint Estimation of 3-Dimensional Parameters

针对信号相位匹配奇异值分解(SVDSPM)算法中参数联合估计耗时长的问题,提出了免疫记忆量子克隆算法(IMQCA).该优化算法引入模拟退火机制修正量子旋转门函数的旋转角度值,构建记忆单元保留进化历史最佳抗体,并结合克隆算子加速种群收敛.由SVDSPM平面阵算法构造了IMQCA的目标函数,提出了同时估计信号方位角、俯仰角和频率的SVDSPM联合估计算法.仿真结果表明,IMQCA算法的方位估计精度与传统的SVDSPM算法相当,但计算耗时仅约为后者的10%,且低信噪比下的性能优于MUSIC方法.在-10 dB信噪比下,IMQCA所得方位角、俯仰角和频率的标准差分别比标准遗传算法小6.659°、9.645°和28.634 Hz,比量子免疫克隆算法小0.789°、1.075°和0.864 Hz.

Focusing on the computational load of estimating parameters in the algorithm of singular value decomposition based on the signal phase matching principle （SVDSPM）, a novel immune memory based quantum clone algorithm （IMQCA） is proposed to optimize the searching procedure of SVDSPM algorithm. Antibodies in a population are represented by quantum bits, and the quantum rotation gate strategy and a dynamic adjusting rotation angle mechanism based on simulated annealing are applied to accelerate convergence with the clone operator. The antibody population is used for global search and an immune memory unit is set to reserve the best antibody. Furthermore, an objective function of IMQCA with three independent variables is derived by the plane array algorithm of SVDSPM. Simulation results show that the proposed algorithm is effective to reduce the computational time-consumption, and outperforms the MUSIC method in low SNR. Compared with the genetic algorithm and quantum-inspired immune clone algorithm, the performance of IMQCA is better.