Design Of Distributed Compressed Sensing System Based On Wireless Sensor Network

Distributed compression sensing technology is a novel video application solution,which shifted the computing complexity from the video encoding to the decoding side and have a broad application prospects in the mobile video communications,environment monitoring and disaster relief.This paper focus on the design and implement a distributed compression sensing system based on wireless sensor networks.Firstly,according to the requirements of environment monitoring and emergency rescue,this paper combines the distributed compressed sensing and wireless sensor network technology to analyze the demand and performance of the system.The overall architecture of the system is designed,the software and hardware implementation scheme of each functional module is introduced in detail,the data acquisition and coding of the sensing node is realized,Ad Hoc network is used to carry out the data transmission.The video decoding reconstruction is conducted in back end.Secondly,in view of the quantization process of compressed sensing,this paper considers the correlation of the observed values,and applies the image quantization algorithm to the distributed video compression system.The observations are stratified according to correlation,observations of keyframe values generate side information to complement the non key frames,which further reduce encoding complexity.According to the distribution characteristics of the observations,a Gauss random quantization matrix is constructed.The experimental results show that the improved algorithm can improve the rate distortion performance of the system,and the Gauss random quantization matrix can improve the reconstruction quality of CS frame.Finally,the paper constructs and implements the sensing node and platform.The function and performance of the system are tested by the distributed video compression and quantization algorithm.The results show that the distributed compressed sensing system based on wireless sensor network can meet the practical needs of the field and emergency rescue environment.