Applied And Research In Wireless Sensor Networks With Kalman Filter Localization Algorithm

Wireless sensor networks are self-organizing networks. They consist of a largenumber of sensor nodes with sensing, computing and communication capabilities, havebeen widely used in many fields. Based on research and analysis a number of importantwireless sensor network node localization algorithms, and analyzed questions exist inapplications in practical in positioning algorithm of current issues. This paper proposeda new location algorithm, in the present wireless sensor network nodes configurationcases, improve the positioning accuracy of wireless sensor network nodes. the core ofthis article is surrounding the wireless sensor node localization algorithm, withoutincreasing the computational overhead in premise, to achieve a improvement inpositioning accuracy. The following outlines the main content and innovation of thispaper:1. Summarized the existing wireless sensor network node localization algorithms,including the composition and characteristics of wireless sensor network (WSN), keytechnologies and localization performance index and influencing factors. Introduces theprinciple and performance evaluation index of some typical wireless sensor nodelocalization algorithm exhaustively, and carried on the comprehensive analysis of thesetypical localization algorithm, sums up the deficiencies and problems need to besolved in the localization algorithm in actual applications.2. This paper has researched and analysised SDSTWR (Symmetrical Double-SidedTwo Way Ranging, namely bi-sided symmetrical ranging France) ranging algorithmbased on CSS (chirp spread spectrum technology)，insides, verificate the algorithm hassmall ranging deviation, wide ranging scope,low cost and suitable for indoor andoutdoor environments through experiments.In considering that in actual positioningscenario, there are many unknown presence of noise impact on the positioning accuracy,we proposed a distance-based SDSTWR weighted maximum likelihood estimation andKalman filtering node localization algorithm. Firstly, measure the distance between theunknown node and anchor nodes by SDSTWR,then using WMLE (weighted maximumlikelihood estimation) converted a non-linear distance measurements value into targetstate’s observed value, and finally have iteration of target state observations throughKalman filter algorithm, to solve the effects of positioning accuracy caused by noise.3. Designed the system simulation for the algorithm based on SDSTWR rangingweighted maximum likelihood estimation and kalman filter proposed in this paper, anddesigned and implemented the wireless sensor network positioning system4. Has carried on the simulation analysis of the proposed algorithm andImplementation of positioning system. The experiments including the construction ofthe simulation environment, ranging Data collection; Analyzed and compared thetraditional maximum likelihood estimation algorithm and the positioning algorithmproposed in this paper(based on SDSTWR distance weighted maximum likelihood estimation and KALMAN filter).through the experiment analysis, the positioningalgorithm proposed in this paper wins a higher positioning accuracy and a lowerpositioning deviation.