Genetic neighbor set construction method for assisting mobile node position anomaly detection

Abstract

The invention relates to a good neighbor set construction method for assisting mobile node position anomaly detection, which can obtain the specific area of a node through area division according to the position data of the mobile node in each time slice, and further discover the current neighbor relation of the node. On the basis, a current good neighbor set of the node can be constructed according to a plurality of neighbor relations of the node under a continuous time slice with a certain length, meanwhile, the neighbor number mean value of the node is counted, and finally, whether the position information of the mobile node in the current time slice is abnormal or not is judged according to the good neighbor set and the neighbor number mean value. The method can be applied to traffic monitoring of common roads, tunnels and bridges in smart cities.

Description

technical field [0001] The invention relates to the technical field of computer data mining and analysis, in particular to the technical field of related stream data mining and analysis, and in particular to a good neighbor set construction method for assisting mobile node location anomaly detection. Background technique [0002] With the increasing popularity of applications such as GPS positioning, sensor networks, and wireless communications, more and more trajectory data are collected and saved to application servers. How to find abnormal location information from these trajectory data sets has gradually become an interesting research. subject. [0003] Commonly used trajectory location anomaly detection methods include: using a distance function to calculate the distance between trajectories to detect abnormal trajectories, or classifying trajectories and using the idea of ​​a classifier to detect abnormal trajectories. The anomaly detection methods disclosed in the ex...

AI Technical Summary

Problems solved by technology

And CN108242145A provides a track point detection method and device, the patent calculates the difference between the two according to the real curvature and driving curvature corresponding to each sampling point on the vehicle track, and compares the difference with the preset curvature difference threshold Compare and determine abnormal trajectory points; this method can accurately detect abnormal trajectory points, but this method uses peripheral sampling devices. This passive sampling method is not suitable for the actual situation of large traffic flow, and requires additional costs.

Images