论文部分内容阅读
传统的盲辨识算法是基于通道输入的统计模型和通道输出 ,然而在对输入的统计特性进行准确估计时需要大量的数据。为了避免这种缺点 ,本文提出了一种单输入多输出 (SIMO)线性时不变有限单位冲击响应 (FIR)系统的盲辨识最小二乘算法 ,利用递推最小二乘算法求解这类算法中的 XL TK方程 ,大大降低了算法对计算存储量的要求。在盲辨识的基础上 ,利用多项式互质的一个判别定理 ,通过解卷积求出 SIMO- FIR系统输入 ,最后通过实验验证了算法的有效性。
Traditional blind identification algorithms are based on statistical models of input channels and channel outputs, but require large amounts of data to accurately estimate the statistical properties of inputs. In order to avoid this shortcoming, this paper proposes a blind-identification least-squares algorithm for single-input multiple-output (SIMO) linear invariant finite impulse response (FIR) systems. Recursive least squares XL TK equation, greatly reducing the algorithm for calculating the storage capacity requirements. On the basis of blind identification, the SIMO-FIR system input is obtained by deconvolution using a discriminant theorem of polynomial coprime. Finally, the validity of the algorithm is verified through experiments.