Reach On Energy Saving Strategies On WSNs Based On The Sptial-temporal Correlations

The addition of a wireless sensor network technology to the transportation facilities of Intelligent Transportation System(ITS)will fundamentally alleviate the problems of safety,smoothness,energy saving and environmental protection that plague modern traffic,while also improving the efficiency of transport.Wireless Sensor Network(WSN)obtains environmental information through wireless sensor nodes,which can communicate with each other and automatically construct network by self-organized way.WSN is widely used in data acquisition,pollution monitoring and disaster prevention.As the sensor nodes work in special environment,and they are powered by battery,so the energy efficient strategies are regarded as a key technology of WSN protocol research.The periodically collected data of single sensor in WSN are temporal correlated,and they are also spatial correlation if the nodes are close to each other.Therefore,redundant data can be removed by some transformation to compress data and save energy.The paper is focused on the study of energy saving strategies based on temporal and spatial correlation.Firstly,based on the correlation analysis of periodic data,the study of the relations between the sampling frequency and data distortion,and the comparison between the Lossy Compressionand Lossless Compression,an improved LTC algorithm is proposed,which is the trade-off between reconstruction accuracy and sampling frequency based on the threshold.Simulation result show that the proposed algorithm can suppress the frequency,reduce the redundancy,save energy and restrain the distortion at the same time.Secondly,the influences of spatial correlation on the distortion are discussed,such as distance between nodes and channel attenuation.It prove that the representative nodes can be selected to reduce the distortions of the whole network under the condition of the data are spatial correlated.The concept of the relevant radius is proposed to build the relevant clusters.According to the correlation coefficient and the location information of the nodes,the distortion function is determined,and the cluster head nodes are selected rationally.The spatial correlation between the data is not only effectively used to ensure the distortion within a certain range,but also avoid transmitting too much data and exhausting energy.The GCC and K-Means algorithms are further analyzed,Both of the algorithms can effectively suppress data traffic,reduce data redundancy,and optimize network power consumption.and the results show that K-Meansalgorithm can get more uniform clusters and a smaller average distortion in the case of the same area and number of nodes.Finally,using the results of correlation analysis,K-Means and GCC(Greedy Corrected Clustering,GCC)algorithm are applied to the typical LEACH(Low Energy Adaptive Clustering Hierarchy)protocol,the simulation results show that the,In different scenarios,different algorithms can be applied to different needs,fusion of these algorithms can further save energy,improve data accuracy,reduce distortion,and prolong network lifetime.