论文部分内容阅读
现有二维到达角估计算法大多基于子空间理论及需要参数配对,针对这一问题,在稀疏表示理论框架下提出了一种参数自动配对的二维到达角估计新算法。该算法在L阵列下构建阵列互相关矩阵的稀疏表示模型,利用奇异值分解降低复杂度并基于群LASSO(Least Absolute Shrinkage and Selection Operator)获得方位角估计。在方位角估计的基础上,基于向量化操作构建稀疏空间谱匹配模型,然后利用LASSO获得俯仰角估计。与参数配对ESPRIT和改进的传播算子方法相比,所提算法不仅无需参数配对过程,而且可以提供改进的估计精度。计算机仿真结果验证了所提算法的有效性。
The existing 2D DOA estimation algorithms are mainly based on the subspace theory and the required parameter matching. To solve this problem, a new two-dimensional DOA estimation algorithm based on sparse representation theory is proposed. The algorithm constructs a sparse representation model of array cross-correlation matrix under L-array, reduces the complexity by using singular value decomposition and obtains the azimuth estimation based on LASSO (Least Absolute Shrinkage and Selection Operator). Based on the azimuth estimation, a sparse spatial spectrum matching model is constructed based on vectorization operation, and then the pitch angle estimation is obtained by LASSO. Compared with parameter matching ESPRIT and improved propagation operator method, the proposed algorithm not only does not need parameter matching process, but also can provide improved estimation accuracy. Computer simulation results verify the effectiveness of the proposed algorithm.