Variational Bayesian Expectation-Maximization Positioning Method and System for Wireless Sensor Networks

Abstract

The invention discloses a wireless sensor network variational Bayesian expectation maximization positioning method and system. The method includes: positioning a wireless sensor network containing holes to obtain positioning information; determining the location on the boundary of the hole according to the positioning information node; adopt multi-hop mode to obtain the distance information between each of the nodes; determine the knuckle information of each of the nodes; according to the distance information and the knuckle information, establish a maximum likelihood estimation model with hidden variables; according to the The above maximum likelihood estimation model establishes a variational Bayesian expectation maximization positioning model; updates the parameters of the variational Bayesian expectation maximization positioning model, adjusts the estimated position of the unknown node, and after multiple iterations, the unknown node is obtained location information. The positioning accuracy of the wireless sensor network can be improved by adopting the invention.

Description

technical field [0001] The invention relates to the field of wireless sensor network node positioning, in particular to a wireless sensor network variational Bayesian expectation maximization positioning method and system. Background technique [0002] Wireless sensor network (wireless sensor networks, WSNs) node positioning technology is an important supporting technology to realize advanced scientific concepts such as big data analysis and Internet of Things applications. Therefore, it is of great significance to use wireless sensor network information to locate sensor nodes. At present, the research on positioning technology for uniformly distributed wireless sensor networks is relatively mature. The basic positioning methods include DV-HOP, centroid algorithm, APIT and other non-ranging methods, as well as TOA, AOA and other ranging methods. Based on the above methods, , some more accurate and complex methods including collaborative methods, distributed methods, convex ...

AI Technical Summary

Problems solved by technology

[0004] However, compared with the requirements of wireless sensor networks in practical applications, the positioning accuracy of existing methods cannot meet the standard, and the large deviation of positioning results will cause serious obstacles to data processing and decision-making execution, which is not conducive to wireless sensor networks. Application in practical engineering

In addition, existing methods have high requirements on sensor hardware, communication conditions, and node computing capabilities, which increases the difficulty of positioning large-scale wireless sensor networks in complex environments.

Images