虚拟阵列Khatri-Rao积与子空间联合稀疏表示的DOA估计方法

Virtual Array Khatri-Rao Product and Subspace Joint Sparse Representation Method of DOA Estimation

摘要:

利用目标信号在空域分布的稀疏性,该文提出了一种基于虚拟阵列Khatri-Rao(KR)积与信号子空间联合稀疏表示的单快拍DOA估计方法；该方法利用单次快拍的采样数据,构造出双向虚拟阵列数据,并对虚拟阵列数据的协方差矩阵进行KR积变换处理,然后对向量化后的数据进行顺序重构,利用重构矩阵的大奇异值对应的左奇异向量为估计信号子空间；最后,利用凸优化工具箱对稀疏模型进行二阶凸规划的优化求解,得到高精度的DOA估计值；仿真实验验证了算法的有效性,在低信噪比下比传统MUSIC和OMP算法具有更高的估计精度。

Abstract:

Using the target signal in the spatial distribution of sparse, this paper puts forward a Khatri-Rao(KR) product based on virtual array and signal subspace joint sparse representation of single snapshot DOA estimation method. The method uses a single snapshot sampling data, constructs the two-way virtual array data, and the covariance matrix of the virtual array data for KR product transformation process, and then to reconstruct the order of data after vectorization, by using the large singular values of reconstruction matrix left singular vectors of the corresponding to estimate the signal subspace; Finally, using convex optimization toolbox for sparse matrix model of quadratic convex programming optimization solution, get high accuracy DOA estimate. Simulation experiments verify the effectiveness of the algorithm, under the low SNR has higher estimation accuracy than traditional MUSIC,SVD and OMP algorithm.