Method for compensating for low-pass filter of compressed sampling system

Abstract

The invention provides a method for compensating for a low-pass filter of a compressed sampling system. The method includes the steps that a sampling value error ye of the actual low-pass filter of the compressed sampling system is obtained by inputting a known tested signal x (t), a nonideal low-pass filter FIR coefficient error e is estimated, and then a nonideal low-pass filter FIR coefficient h is obtained; an FIR coefficient hd for correcting the filter is estimated on the basis, and then the corrected nonideal low-pass filter is made to be turned into an ideal low-pass filter; meanwhile, sampling frequency of the compressed sampling system is increased, the oversampling coefficient of the compressed sampling system is R, sampling values of channels are sampled, and then the sampled sampling values are filtered through a compensating filter. It is guaranteed that the sampled and filtered sampling values are equal to sampling values obtained through the ideal low-pass filter, and therefore the problem that large errors exist in the reconstruction process of a tested signal sequence x* (n) caused by the nonideal low-pass filter in the compressed sampling system is solved.

Description

technical field [0001] The invention belongs to the technical field of spectrum sparse signal compression sampling, and more specifically relates to a method for compensating the low-pass filter of the compression sampling system, which is used to compensate the non-ideal characteristics of the low-pass filter to reduce the compression sampling rate. The difficulty of system design. Background technique [0002] Compressive sampling technology based on modulated broadband converter is an under-sampling method based on compressed sensing theory, which can not only break through the limitation of sampling theorem on the sampling frequency of analog-to-digital converter (ADC), but also avoid the impact of ADC analog signal input bandwidth. The limitation of the sampling signal frequency. [0003] The existing relatively mature compressed sampling system uses m-channel pseudo-random sequence p i (t), i=1, 2,..., m sense the spectrum of the measured signal x(t) (mix in the mixe...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to overcome the deficiencies in the prior art, to provide a method for compensating the low-pass filter of the compressed sampling system, to solve the problem of reconstructing the measured signal sequence x caused by the non-ideal low-pass filter in the compressed sampling system * [n] There is a large error problem

Images