Application Research On Compressive Sensing In WSN Data Collection

As one of the important components of the Internet of Things and big data,WSN(Wireless Sensor NetworK)has been widely used in various fields such as environmental monitoring and intelligent transportation.In order to fully sense the physical information,nodes are usually redundantly deployed in the monitoring area with the repeated data collection.For sensor nodes with limited processing capacity,there are excessive data transfer and higher redundancy.At the same time,energy consumption will significantly increase when using the batteries power supply.The scale and stability of the WSN will be hindered.By using CS(Compression Sensing)in WSN,data compression and collection can be performed simultaneously.It will effectively reduce the amount of data transmission and networK energy consumption,and it will improve system stability.This paper does the following worK.(1)Application analysis and theoretical study of CS in WSN data collection.Combined with the WSN data collection,research the effects of data collection on the amount of node data transmission when using the CS algorithm.The common sparse basis and observation matrix are studied experimentally,and the influence of data volume and compression ratio on the data reconstruction effect is analyzed.(2)A dynamic data compression collection scheme based on LEACH(Low Energy Adaptive Clustering Hierarchy)routing protocol and CS algorithm CS-LEACH-I is proposed.First of all,In order to solve the problem of uneven distribution of nodes,uneven energy consumption,and simultaneous data collection and compression,proposing a LEACH-I algorithm.Then,designing CS-LEACH-I which combines the algorithm in the clustering routing process to achieve data collection and compression.Finally,experiments are carried out from two aspects of energy consumption and data recovery.(3)Research on adaptive ISAMP reconstruction algorithm.An improved SAMP Optimization Algorithm—ISAMP is proposed.For the inconsistent clustering each round and unKnown sparseness of collected data.The sparsity adaptability is optimized by improving the choice of atomic support sets.The good reconstruction effect and higher stability of the algorithm are verified by a large of experiments.