The Research Of The Data Reconstruction In Wireless Sensor Networks

The integrity of the data collection is an important issue for the applications in wireless sensor networks.However,data loss is one of widespread phenomenon in actual data collection,constraining the scope and effectiveness of the applications.At present,the mainstream data reconstruction algorithms often base on the spatial correlation between the sensory data,using interpolation methods for data reconstruction.This method has two main problems:First,as the data loss rate increases,the spatial correlation between the sensing nodes will reduce,thus reducing the accuracy of data reconstruction.Secondly,the sensory data not only have spatial correlation but also have temporal correlation and other characteristics to be further study.In this paper,we first study the data loss patterns of networks,which can be divided into two kinds of specific cases,namely element random loss and block random loss.On this basis,we analyse and model the spatio-temporal correlation of the sensory data,and by this to build the method of data correlation.The main contributions of this dissertation are indicated below.First,we study the problem of data reconstruction in wireless sensor networks,which is based on the surface fitting method.Aiming at the disadvantages of the present data recovery algorithms based on space relativity,we explore the spatial-temporal correlation among the sensory data,and propose a novel data reconstruction algorithm to improve the data error ratio in the wireless sensor networks.The proposed algorithm can utilize the spatial correlation by using the curved face reconstruction,and it is also helpful to improve the accuracy of data reconstruction by exploring the temporal correlation among the sensory data.The experiments demonstrate that the proposed data reconstruction algorithm can significantly reduce the fitting data error ratio compared with related works especially in case that data loss is serious in the network.Second,we explore the problem of data reconstruction in wireless Sensor Networks,which is based on the Markov Random Field.For the indeterminacy of the sensory data spatial corrlation,the spatial correlation of different physical datasets contains different characteristic.We use the Markov Random Field(MRF)to model the spatial correlation of sensory data.Its model can adjust to the indeterminacy of data related degree.In this paper,we propose a data reconstruction method based on the spatio-temporal correlation,which uses temporal correlations as the bound term to improve the result of data reconstruction.The experimental results indicate that the proposed algorithm has large improvement in indoor lighting data restoration,comparing to the related works.