论文部分内容阅读
本文提出了时序递归LS方法的部分自适应波束形成器结构,可以证明这种结构的算法也收敛到维纳方程解。这种结构不需求解协方差矩阵,不涉及矩阵求逆,具有很快的收敛速度,对于大型阵运算量可降低几个数量级,考虑到空间噪声场的时变性,采用传统的最优条件下的估计方法算法会受到影响,作者采取了渐消记忆的措施,有效地消除了因最优化条件得不到满足时引起的性能降低,而不损失自适应算法的收敛和跟踪的基本性能,时变地跟踪干扰环境并直接得到抵消利余输出。为了验证该方法和结构,最后给出了计算机仿真实验结果。
In this paper, we propose a partial adaptive beamformer structure of time-sequence recursive LS method. It can be proved that the algorithm of this structure also converges to the Wiener equation. This kind of structure does not need to solve the covariance matrix, does not involve the matrix inversion, and has a fast convergence rate. It can reduce orders of magnitude for the large-scale array calculation. Taking into account the time-varying nature of the spatial noise field, the traditional optimal condition The algorithm of the estimation method will be affected. The author adopts the measure of fading memory to effectively eliminate the performance degradation caused when the optimal conditions are not satisfied without losing the basic performance of the convergence and tracking of the adaptive algorithm. When Change tracking interference environment and directly offset the surplus output. In order to verify the method and structure, finally the computer simulation results are given.