Positioning algorithm based on linear embedding optimization in internet of sensor network

摘要

In order to solve such problems as accurate positioning difficulty, positioning data cache accurate-reading dissatisfaction and multidimensional dynamic elastic distance measurement unavailability, a sensor network positioning algorithm based on linear embedding optimization was proposed in the paper. Firstly, neighborhood clustering was implemented for the nodes in a wireless sensor network, namely: the first-order neighborhood nodes and the second-order neighborhood nodes were taken as the network neighborhood. Then, other network nodes were positioned according to neighborhood radius rejection and attraction conditions, and meanwhile the error was minimized through iteration mechanism to finally obtain the accurate position coordinates of the node. The simulation result shows: compared with current widely-applied two-dimensional non-iterative accurate positioning algorithms - RSLM algorithm and SOCP algorithm, the new algorithm can effectively reduce the node position error and improve the positioning accuracy in multidimensional distance measurement through the clustering mechanism for accurate positioning and the error correction method based on iteration mechanism, and can also improve network packet delivery ratio and reduce network control overhead.