基于GPU加速的双向平滑MUSIC相干信号DOA估计算法  

A GPU-Accelerated Forward-Backward Spatial Smoothing MUSIC Algorithm for DOA Estimation of Coherent Signals

作　　者：曹泽坤 怀园园 张昕[1] 邱天[1] CAO Zekun;HUAI Yuanyuan;ZHANG Xin;QIU Tian(Xi'an Electronic Engineering Research Institute,Xi'an 710100)

机构地区：[1]西安电子工程研究所,西安710100

出　　处：《火控雷达技术》2024年第4期31-36,41,共7页Fire Control Radar Technology

摘　　要：针对传统MUSIC算法不能有效分辨相干信号以及其特征值计算量庞大的问题,本文提出一种基于GPU加速的双向平滑MUSIC相干信号DOA估计算法。该算法首先对各子阵协方差矩阵求平均得到双向平滑协方差矩阵;然后对该协方差矩阵进行特征值分解,得到正交的信号和噪声子空间;接着建立算法空间谱函数,进行谱峰搜索,得到DOA估计;最后使用GPU对以上步骤进行加速。仿真结果表明,本文所提基于GPU加速的双向平滑MUSIC算法能快速实现相干信号DOA估计。To address the issue that traditional MUSIC algorithms cannot effectively resolve coherent signals as well as their large computational burden related to eigenvalue calculations,this paper proposes a GPU-accelerated forward-backward spatial smoothing MUSIC algorithm for DOA estimation of coherent signals.The algorithm first derives the forward-backward smoothed covariance matrix by averaging the covariance matrices of subarrays.It then performs eigenvalue decomposition on the obtained covariance matrix to derive the orthogonal signal and noise subspaces.Subsequently,a spatial spectrum function is constructed,and spectral peak search is conducted to obtain the DOA estimation.Finally,the above steps are accelerated using GPU.The simulation results show that the proposed GPU-accelerated forward-backward spatial smoothing MUSIC algorithm can rapidly complete DOA estimation of coherent signals.

关 键 词：相干信号 DOA估计 双向平滑MUSIC算法 GPU加速 

分 类 号：TN953[电子电信—信号与信息处理]