Research And Design Of Analog-to-Information Convertor Based On Compressed Sensing

Compressed Sensing(CS)is a new theory and method of signals sampling emerged in recent years,which has attracted the attention of researchers in many fields.CS has great theoretical values and potential application prospects in wireless sensor network.For example,instead of traditional sampling devices,CS theory can reduce energy required in signal sampling,storage,transmission so that it can prolongate battery life of wireless sensor network.In this thesis,an improved Analog-to-Information Convertor(AIC)is proposed based on CS theory and Random Demodulation model.Experimental verification has been completed by the improved AIC which is used for engineering structure monitoring wireless sensor network.The main research contents are as follows.Firstly,the framework of CS theory from signal sparse expression,measurement matrixes and signal reconstruction are introduced.The research focuses on Orthogonal Matching Pursuit(OMP)algorithm.The advantages and disadvantages of OMP algorithm has been summed up by combining different reconstruction results for sparse signals in this thesis.Secondly,an improved AIC hardware system is designed which is centered on FPGA.The improved AIC hardware system has been improved from 3 parts: signal acquisition,compressed sensing and communication.Finally,based on the improved AIC system,a wireless sensor network system for engineering structure monitoring is set up,and the signal experiment on 3 layers metal frame and Bingzhou Yellow River Bridge model has been completed.The experimental results verify the feasibility of CS theory in wireless sensor network applications on the one hand and reveal how the overall performance of the system affected by different design of analog information converter module on the other hand.