Research On High Energy Efficiency Data Collection Method Based On Spatio-temporal Correlation

For the past few years,the increasing maturity of communication,sensing and other technology has pushed forward the continuous improvement of the influence of wireless sensor networks(WSNs).It extends the Internet from the virtual world to the physical world.In order to realize the intelligent communication of various things and meet the increasing requirements of people's production and life,the Internet of Things(Io T)supported by WSNs was born.The progress of WSNs and the Io T promote the development of monitoring applications based on them,which makes monitoring applications play an important role in human life.Since the nodes in non-rechargeable WSNs only rely on batteries for energy,limited by the size of nodes and the environment of monitoring area,the energy stored by nodes is less,and it is difficult to replace the batteries for the nodes.Therefore,how to extend the lifetime of WSNs is a problem that people have been paying attention to.Due to the spatio-temporal correlation of nodes in the WSN,nodes in the network will transmit a lot of similar data,which wastes a lot of energy of nodes.Most of the current work extends the lifetime of the network by using the temporal or spatial correlation of the network.But in some of these methods,only the temporal or spatial correlation of nodes is considered separately.The other part considers the spatio-temporal correlation of nodes at the same time,but there are still problems in the implementation of high computational complexity and difficult to determine parameters.Based on the above reasons,this paper considers the spatio-temporal correlation of data and proposes an approximate data collection method(STAC).This method puts forward the concept of correlation-variation among nodes when recovering the data of sleep nodes,which makes recovery data more accurate.Within the range of error tolerance,STAC reduces the collection and transmission of redundant data in the network from the source,and extends the lifetime of the network.In spatial,STAC combines the spatial recovery function,correlation-variation verification mechanism and the selection mechanism of work nodes proposed in this paper,which ensures the minimum number of work nodes in each work cycle.When the working cycle is over,recover the data of sleep nodes under the premise of ensuring the accuracy of the recovery data.In temporary,STAC adopts the method of dynamically adjusting the sampling interval of work nodes to further reduce the amount of redundant data collected by work nodes.After the end of the work cycle,the time recovery function mentioned in this article is further used to recover the data that was not collected due to the adjustment of the sampling interval within the tolerable error range.Finally,this paper uses real data to do a lot of experiments,and the experimental results prove the advantages of the STAC in extending the network life cycle.The experimental results show that when the predefined error is 0.4,the initial energy of nodes is 0.2J and the threshold of data difference between two consecutive samples of the node is0.05,compared with the lifetime of the network using OPR or OIQL,the lifetime of the network using STAC is extended by 414% and 34.555% respectively.In particular,when comparison methods and STAC are limited by temporal or spatial,STAC can still extend the network life cycle better.