简化正交归一约束的自适应噪声子空间估计算法

为避免MUSIC算法的特征分解过程,提出一种噪声子空间的自适应估计算法,能够估计整个噪声子空间.该算法基于正交归一化约束的最小均方(LMS)算法,但对正交归一约束过程进行了简化,较之显式正交归一化约束的LMS算法,简化了运算过程,适合实时计算与工程实现.噪声子空间估计以迭代的方式进行,适合应用于运动信号源的跟踪.仿真结果显示算法具有很好的空间谱估计性能和DOA跟踪性能.

Symplified Orthonormalization Constraint Adaptive Noise Subspace Estimation Algorithms

Most of the computational complexities of MUSIC algorithms are centered in eigendecomposition,to avoid eigendecomposition,a novel adaptive noise subspace estimation algorithm is provided.The method is based on orthonormalization constrained LMS algorithm,but the orthonormalization operation is symplified and the computational complexities is reduced greatly in contrast to explicit orthonormalization operation.The algorithm suits egnieering implementation and work iteratively which can be used for real-time DOA tracking.The simulation analysis indicates that this algorithm is active.