Research On Sub-Nyquist Sampling Based On Modulated Wideband Converter

Compressive sensing is a theory that can sample sparse signals without distortion when the rate is far below the Nyqvist sampling rate.It utilizes the characteristics of sparse signals to sample and reconstruct signals at an extremely low rate,reduce the sampling rate required for signal transmission systems,and reduce the transmission and storage pressure of signal processing equipment.The multi-band signal is ubiquitous in radar and communications systems.It usually consists of multiple high-carrier narrowband signals,so its spectral range is wide and its frequency domain is sparse.Due to the wide spectral range of multi-band signals,it is difficult to do Nyquist sampling of the ADCs.Modulated wideband converter(MWC)is based on the theory of compressive sensing.It is an ingenious solution to the multi-band signal sampling problem.Besides,the reconstruction algorithm is a key factor in determining the performance of MWC system.The main research work of this thesis is as follows:1.This thesis researches the compressive sensing theory and the analog information conversion technology.Firstly,this thesis introduces the basic principle of compressive sensing theory,including the observation matrix,the sparse basis and the reconstruction algorithm.Secondly,this thesis analyses the analog-to-information conversion technology,and deeply understands the principle of compressed sampling of the modulation broadband converter.2.Based on the limitations of existing reconstruction algorithms which need to know the signal sparsity,combining with the existing algorithms,this thesis designs a multi-band signal compression sampling and reconstruction scheme on account of the MWC system,sets up a corresponding signal processing system,and then comes up with the Iterative Support Detection Multiple Measurement Vectors(ISDMMV).The experimental results show that the compressive sampling signal processing system based on the MWC system has a good performance,and it could balance the performance of the algorithm with the factor of without priori information.It breaks the limitations of the existing conditions,and achieves the goal of this thesis.3.This thesis points out that ISDMMV algorithm still has room for improvement.Therefore,this thesis applies the support refining to the ISDMMV algorithm.and proposes an improved algorithm named SR-ISDMMV.The experimental results show that the SR-ISDMMV algorithm could improve the performance of ISDMMV effectively.It improves the success rate of MWC system when the sampling compression rate and SNR are low,and then improves its performance when the sampling compression rate and SNR are high.