Research On The Improved Localization Algorithm Of Mobile Wireless Sensor Based On Monte Carlo

Wireless sensor networks(WSNs)have been the focus of research in recent years,among which node localization technology is one of the key technologies in WSNs.At present,scholars’ research achievements on static node localization are abundant,but research on mobile node localization is relatively lacking.With the application environment of wireless sensor network becoming more and more complex,mobile node localization becomes more and more important.Compared with traditional mobile node localization,Monte Carlo mobile node localization algorithm makes use of the mobility of nodes and overcomes the influence of node motion speed on localization accuracy.However,There are some problems in Monte Carlo node localization algorithm,such as too much dependence on the number of anchor nodes,low sampling efficiency,low localization accuracy and low localization coverage.To solve these problems,considering that unknown nodes and anchor nodes in wireless sensor network are in random movement,this paper studies the improvement of mobile node positioning algorithm based on Monte Carlo.The main contents are as follows:1.Aiming at the problems existing in the RSSI-based Monte Carlo node localization algorithm,such as high demand for anchor nodes,high localization cost and low sampling efficiency,a Monte Carlo node localization algorithm(TANRMCL)based on temporary anchor nodes was proposed.In this algorithm,an appropriate common node is selected as the temporary anchor node.the anchor node and the temporary anchor node is used as the constraint condition to construct the sampling area.Then the correction radius is used to modify the localization information of the temporary anchor node to reduce the error.Finally,the improved resampling method is used to improve the sampling efficiency.The simulation results show that this algorithm improves the sampling efficiency of nodes,reduces the density of anchor nodes in the network,thus reducing the cost of the whole network,and also improves the positioning accuracy and coverage of nodes to some extent.2.For scenarios with high positioning accuracy requirements,an IQUATRERMCL node localization algorithm is proposed based on an improved QUasiAffine TRansformation Evolutionary.The IQUATRE-RMCL algorithm chooses an affine transform evolutionary algorithm(QUATRE)with strong optimization capability as a basic algorithm to improve the positioning algorithm,and then puts forward an improved QUATRE algorithm(IQUATRE)to solve the problem of too slow convergence and easy to fall into local optimum.The IQUATRE algorithm firstly uses the reverse learning strategy to generate more uniform initialization particles.Then,the value of the confidence function of the particles was compared in pairs.The winning particles were added to the dominant population and the losing particles were added to the inferior population.Different evolutionary strategies were selected for the two populations.Finally,the adaptive scale factor is proposed.This method improves the above problems of QUATRE algorithm to some extent.The weighted average value of IQUATRE instead of TANRMCL algorithm is used to calculate the stage of unknown node localization,and the distance between unknown node and anchor node and temporary anchor node is used as the confidence function to obtain the optimal node position in the sampling area.The simulation results show that the algorithm improves the location accuracy of nodes and the location coverage of nodes.