Reserch On Distributed Location Algorithm Of Wireless Sensor Network

In the applications of wireless sensor networks, the location information takes animportant role, and the determination of the nodes’location is an important part of thistechnology. So, the research of effective positioning algorithm has importanttheoretical significance and pragmatic value. This paper proceeded from in-depthanalysis of algorithm based on multidimensional scaling, studied the reason of its highcomplexity, and then improved the location algorithm.First of all, based on the research of plenty of related literatures, background andsignificance of wireless sensor networks were summarized. This paper described theindex of performance of location algorithm, determined the direction of improving,simulated and analyzed the existing typical algorithms.Secondly, this paper studied the principles of multidimensional scalingtechniques. If we regard the distance between the nodes as high-dimensional data,regard coordinates of the nodes as low-dimensional data, through the process ofdimension reduction can convert the distance into coordinate information. And then,this paper analyzed in detail and implemented the various algorithms based onmultidimensional scaling techniques technology, defined the reason for highcomplexity of this algorithm. It used all of the distance information between nodes todetermine the location of the nodes, this increased the computational complexitysignificantly. Focused on this problem, this paper proposed to use another datadimension reduction tool, which called stochastic proximity embedding instead ofmultidimensional scaling techniques, and then this paper got an improved locationalgorithm that had high positioning accuracy and low computational complexity.Finally, this paper described the implementation steps of the improved locationalgorithm, determined the parameters of practical applications, and this papercompleted the algorithm simulation and performed comparison using MATLABgraphics data processing software. Simulation results showed that the improvedalgorithm achieved the purpose of reducing the computational complexity.