基于四阶差分共阵和遗传算法联合优化的MUSIC算法  

MUSIC Algorithm Based on Joint Optimization of Fourth-order Difference Co-array and Genetic Algorithm

作　　者：胡连杰 于勇 赵艳秋 刘一帆 HU Lianjie;YU Yong;ZHAO Yanqiu;LIU Yifan(The Eighth Research Academy of CSSC,Yangzhou 225101,China)

机构地区：[1]中国船舶集团有限公司第八研究院,江苏扬州225101

出　　处：《舰船电子对抗》2025年第2期46-51,57,共7页Shipboard Electronic Countermeasure

摘　　要：为了解决基于二阶累积量的多重信号分类算法(MUSIC)在处理非高斯信号时存在的高阶信息缺失和实时性不足问题,提出了一种基于四阶差分共阵与改进的遗传算法的联合优化方法。结合四阶累积量抑制高斯噪声和扩展虚拟阵元的优良性能,首先构造嵌套阵的四阶差分共阵(FODC),将M个物理传感器扩展到O(M^(4))个虚拟单元,然后与矢量化的四阶累积量矩阵相匹配得到单快拍向量,再结合空间平滑得到MUSIC空间谱。最后引入改进的遗传算法,通过快速非支配排序和自适应交叉变异联合策略,在保留全局搜索能力的同时,实现多目标波达方向估计。通过仿真实验验证,新算法有效提高了面向非高斯信号波达方向估计的测量精度和实时性。To address the issues of higher-order information loss and insufficient real-time performance in processing non-Gaussian signals by using second-order cumulant-based multiple signal classification(MUSIC)algorithm,a joint optimization method based on fourth-order difference co-array and improved genetic algorithm is proposed.Combining the superior performance of fourth-order cumulants in suppressing Gaussian noise and expanding virtual array elements,this paper firstly constructs the fourth-order difference co-array(FODC)of a nested array,extends M physical sensors to O(M^(4))virtual units,then matches it to the vectorized fourth-order cumulant matrix to obtain a single-snapshot vector,which is combined with spatial smoothing to derive the MUSIC spatial spectrum.Finally,an improved genetic algorithm is introduced,and multi-objective direction-of-arrival(DOA)estimation is achieved while preserving global search capabilities by incorporating fast non-dominated sorting and an adaptive crossover-mutation joint strategy.Simulation experiments demonstrate that the proposed algorithm significantly enhances the measurement accuracy and real-time performance for DOA estimation of non-Gaussian signals.

关 键 词：波达方向估计 非高斯信号 空间平滑MUSIC 嵌套线阵 改进的遗传算法 

分 类 号：TN911.7[电子电信—通信与信息系统]