|
【文件名】:06619@52RD_HOS.rar
【格 式】:rar
【大 小】:8K
【简 介】:Reconstruction of the impulse response of an unknown system, whose output is observed in the presence of Gaussian noise of any covariance, is based on the use of a pair of 1-D HOS slices. As long as the Hadamard distance between the slices, and the grid size used to discretize the n-th order spectrum are coprime integers, the unknown system can be uniquely recovered, within a complex constant and a circular shift.
Assuming that a pair of 1-D n-th order spectra slices has been chosen, a determined system of linear equations, of the form A x = c, can be formed based on the difference between the logarithms of the samples along the two slices. The unknown vector x consists of samples of the logarithm of the discrete frequency response of the unknown system. Given the Hadamard distance r between the slices and the discretization grid size, N , the MATLAB function hos_matrix.m creates the sparse system matrix A. This matrix, along with the n-th order spectrum sequence, the order of the spectrum (either 3 or 4, for practical reasons), the grid size and the distance of the slices are used as the input to the function hos_id.m, which performs the system reconstruction. The last routine uses the auxiliary function modulo.m, which computes the modulo of an integer number.
【目 录】:无目录
|
本帖子中包含更多资源
您需要 登录 才可以下载或查看,没有账号?注册
×
|