An anomaly detection method for wireless sensor networks based on spatio-temporal similarity

Abstract

The invention relates to a wireless sensor network anomaly detection method based on spatio-temporal similarity, comprising: obtaining a representative feature attribute set according to a feature set extraction algorithm; mapping the representative feature attribute set to a two-dimensional visual space and obtaining a visualization data; performing temporal similarity calculation on the visualized data according to the temporal similarity to obtain a temporal similarity data model; performing spatial similarity calculation on the data model to complete abnormal detection of wireless sensor networks based on temporal and spatial similarity. By adopting the technical solution of the present invention, the abnormality detection in the sensor network is realized, the method for detecting the feature extraction in the abnormality detection, and the effective and high-accuracy abnormality detection for judging the connection between the abnormal feature attributes As a result, a specific and feasible solution is put forward, which has the ability to deal with the anomaly detection of dense nodes, making the sensor network work more accurately in more complex scenarios.

Description

technical field [0001] The invention relates to the technical field of sensor network diagnosis, in particular to a wireless sensor network anomaly detection method based on time-space similarity. Background technique [0002] A sensor network is a wireless network composed of a large number of stationary or moving sensors in a self-organizing and multi-hop manner. Its purpose is to cooperatively perceive, collect, process and transmit detection information of sensing objects in the geographical area covered by the network, and report to users. It provides users with remote monitoring and control in a distributed computing environment. The main components of the sensor network include the sensor node (Sensor Node) and the base station node (Sink Node). Usually, the sensor node forms a communication network through a wireless multi-hop self-address form, and transmits the collected data back to the base station. Each sensor node is composed of data acquisition module, data p...

AI Technical Summary

Problems solved by technology

[0004] However, the above-mentioned method of abstracting and summarizing event data using a graph model has the disadvantage that for the abnormality of densely arranged sensor node groups, the disadvantage of anomaly detection performance will decrease exponentially with the increase of the number of nodes; the above-mentioned method based on feature selection Methods The method of detecting wireless sensor networks does not propose a specific and feasible plan for how to judge the relationship between abnormal features and obtain effective and high-accuracy abnormal detection results

Images