Research Of Data Collection Algorithm Based On Compressive Sensing In Wireless Sensor Networks

The wireless sensor network(WSN)is composed of a large number of wireless sensors in a self-organizing manner.Its own structural characteristics make it have the advantages of high sampling accuracy,good environmental adaptability and strong survivability,and thus it is widely used in environmental data collection in military and civil fields.However,the energy-constrained sensors restrict the large-scale and long-term deployment of WSN in practical applications.Therefore,how to reduce data traffic transmitted in the network,reduce the network energy consumption,and prolong the lifetime of network under the premise of ensuring the accuracy of collected data has become the most urgent problem to be solved in WSN.The emergency of compressive sensing(CS)has provided a new approach to solve above problems.The property of simple encoding and complex decoding and good data compression performance of CS exactly satisfy the requirements of data collection in WSN.Thus,applying CS in WSN has important practical value and research significance for achieving high precision,low energy consumption and longtime data collection.Therefore,for the problem of data collection in WSN,this paper carries out the research of data collection algorithm based on CS in WSN to reduce network energy consumption and prolong network lifetime on the basis of the research on the basic theory of WSN and CS.And from the two aspects: compression coding for the space-time related data in the network by CS to reduce data traffic required to be transmitted and designing efficient network routing algorithm to achieve the generation and convergence of observations in a low-cost and energy-balanced manner,two new spatiotemporal compressive data collection algorithms are proposed:(1)Aiming at the problem that the existing algorithm requires huge overhead to aggregate observations to the sink,with the goal of reducing the energy consumption required for data collection in WSN,this paper proposes a spatiotemporal compressive data collection algorithm based on tendency walk by combining the spatiotemporal compression sampling strategy based on CS with the tendency walk routing.The algorithm fully exploits the spatial and temporal correlation of data in the network.Firstly,the time-domain data gathering by each node is independently compressed to reduce data traffic required to be transmitted in the network.Then,tendency walk strategy is used to fuse the compressed data on each node in the spatial domain to generate the observations and make the overhead required to aggregate the observations to the sink less.The simulation results show that compared with the existing RW,RS and ST-RW algorithms,the proposed algorithm can further reduce the energy consumption during data collection by 82%,52% and 26% while ensuring the quality of reconstructed data.(2)Aiming at the huge overhead required to generate observations and the "hot spots" problem caused by simple multi-hop aggregation routing,with the goal of further balancing the energy consumption distribution and prolonging network lifetime,this paper proposes a cluster-based spatiotemporal compressive data collection algorithm by combining CS and clustering.Firstly,this algorithm proposes a spatiotemporal compressive intra-cluster data collection scheme.It determine the intra-cluster sampled nodes in the spatial domain from the aspects of making full use of data correlation,reducing energy consumption during intracluster data collection and balancing intra-cluster energy consumption distribution,and the time-domain data on the sampled nodes is compressed and encoded by CS to complete spatiotemporal compressive intra-cluster data acquisition,thus reducing the overhead required to generate observations and data traffic transmitted in the network;Then,the algorithm proposes a cluster head rotation strategy to solve the problem of uneven energy consumption distribution caused by different intra-cluster node identity;Finally,the algorithm proposes an energy-balanced routing construction method between clusters to balance energy consumption distribution.It constructs inter-cluster routing according to the remaining energy of each cluster head node and the distance between cluster head nodes,so that after each round of data collection,the variance of the remaining energy between cluster head nodes is as small as possible.The simulation results show that compared with the existing same kind of algorithms Cluster HCS and Cluster STCS,the algorithm can effectively reduce the network energy consumption and balance the network energy consumption distribution under the same reconstruction error,thus significantly prolonging the network lifetime.