A method for detecting spatio-temporal anomalies in online river networks

Abstract

The invention discloses a method for detecting an online river network spatio-temporal abnormal event, which comprises the following steps: firstly, using a connected dominating set algorithm to select a backbone node, and only detecting the state of the backbone node in each time stamp sensor network; then, detecting Whether the backbone node has a spatio-temporal anomaly event; if the backbone node has a time anomaly, add the child node of the backbone node to the candidate set for detection; learn the expected state and observed state of the backbone node with the time anomaly through Bayesian network learning Make a comparison to see if it deviates from the spatial relationship obtained by learning. If a spatial anomaly occurs, the space-time anomaly counter is incremented by 1. At the next timestamp, expand the detection range to include the backbone nodes and nodes in the candidate set, and continue to detect whether spatial anomalies or temporal anomalies occur. Finally, when the number of spatiotemporal anomalies of the backbone node and its candidate set nodes reaches the threshold Q, it is considered that a spatiotemporal anomaly event occurs.

Description

technical field [0001] The invention relates to the technical field of wireless sensor network detection, in particular to a method for detecting spatio-temporal abnormal events in a river network wireless sensor network using a decentralized method. Background technique [0002] With the rapid development of sensor technology and the increasing maturity of wireless communication technology, wireless sensor networks are widely used in ecological scientific observation, environmental monitoring, disaster early warning, and national defense and military fields. In these practical applications, automated anomaly event detection functions are essential. Generally speaking, in order to take corresponding remedial measures for network abnormalities as soon as possible, it is necessary to detect all sudden events that affect system applications in real time, including physical phenomena, pollutant monitoring, etc., and then give early warning prompts. Sensors are deployed in the r...

AI Technical Summary

Problems solved by technology

[0005] Purpose of the invention: a method for detecting spatio-temporal anomalies in online river networks, solving the problem that the prior art cannot comprehensively consider the spatio-temporal correlation of network anomalies

Images