Font Size: a A A

Research On Data Prediction Technology Based On Spatial-temporal Correlation In Wireless Sensor Networks

Posted on:2017-01-30Degree:MasterType:Thesis
Country:ChinaCandidate:Z LvFull Text:PDF
GTID:2308330485990002Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
Wireless Sensor networks(WSNs) nodes are limited by their size. The energy of it is limited. It is a main challenge that limited energy supply for the development of WSNs. The data transmission energy consumption accounts main part of the node energy consumption. The gathered data has strong spatial-temporal correlation in large-scale dense deployment of WSNs node, which makes it possible to predict accurately, reduce the amount of data transmission, energy consumption and data transmission delay by using spatial-temporal correlation of data.This paper deeply studies the principle, characteristics and performance index of typical data prediction algorithm. According to the temporal correlation of sensed data of the single WSNs node and the spatial correlation of sensed data of the multiple WSNs node, two effective data prediction algorithms are presented. The main research content of this paper is as follows:(1)The research of spatial correlation data prediction algorithm based on Markov chain. Prediction algorithm employing spatial correlation can obtain accurate results in irregular changing environment. This paper measure the correlation between adjacent nodes using Delaunay triangle adjacent figure, which based on spatial correlation of data in WSNs. The weight coefficient of the spatial correlation is depending on the distance between adjacent nodes. Finally it describe the change of adjacent nodes accurately using Markov chain in WSNs, and puts forward the data prediction algorithm of spatial correlation based on Markov chain for WSNs. Prediction algorithm of spatial correlation is running in sink node and cluster head, which reduced amount of data transmission between sink node and cluster head by using spatial correlation of data.(2) The research of self-adapting temporal Correlation data prediction algorithm. This paper design self-adapt prediction algorithm based on grey model, which using temporal correlation of data in WSNs. Data is decomposed into the sum of linear part and nonlinear part. Data prediction accuracy is improved by using grey model prediction. Adaptive gray prediction model is running in member nodes and the cluster head, which reduced amount of data transmission between nodes and cluster head using temporal correlation of data.The energy-saving effect of two algorithms in WSNs this paper proposed is verified through the simulation experiment. The simulation results show that the proposed algorithm in this paper can effectively improve predicting accuracy comparing with other prediction algorithms, and prolong the lifetime of WSNs.
Keywords/Search Tags:Wireless Sensor Network, Data prediction, Temporal-Spatial correlation, Grey model, Markov chain
PDF Full Text Request
Related items