Designing method for detecting sensor network abnormal data based on space-time correlation

An abnormal data detection and sensor network technology, applied in the field of information communication, can solve problems such as high computational complexity, limited storage resources, and high communication costs, and achieve the effect of maintaining network security

CN107205244AInactive Publication Date: 2017-09-26HARBIN INST OF TECH AT WEIHAI
0 Cites 9 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
Publication Date
2017-09-26
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure 1
    Figure 1
Patent Text Reader

Abstract

The present invention relates to the field of information communication, and discloses a design method for sensor network abnormal data detection based on spatio-temporal correlation, which includes space dimension detection and time dimension detection, and can realize the detection and classification of abnormal data in wireless sensor networks. and classification results, can respond to events in the network in a timely manner, and at the same time, for malicious nodes that affect the observation results of the base station by sending malicious data, thereby reducing the reliability of the network, by reducing their reputation in the network, the data will not Forwarding from malicious nodes, if the reputation of the node is low to a certain extent, the node will be blacklisted and its data will no longer be received, so as to achieve the purpose of shielding such malicious nodes and maintaining network security.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the field of information communication, in particular to a design method for abnormal data detection of a sensor network based on time-space correlation. Background technique

[0002] The abnormal data of wireless sensor network is very important for environmental monitoring. In actual situation, there may be two kinds of abnormal data, which are malicious data and event data. Malicious data can reduce network reliability by affecting the observation results of the base station, while event data is an important manifestation of environmental changes. Event data can be used to obtain changes in the monitoring area. How to accurately identify abnormal data and effectively distinguish them, and how to accurately understand the changes in the monitoring area while maintaining network security is a current research hotspot.

[0003] At present, the abnormal data detection methods that are widely used are mainly based on statistics ...

Examples

Embodiment 1

[0025] Such as figure 1 As shown, the present invention discloses a design method for sensor network abnormal data detection based on spatio-temporal correlation. The abnormal data detection includes spatial dimension detection and time dimension detection, and the spatial dimension detection is a spatial anomaly based on the K-Means clustering method Data judgment, the time dimension detection is based on the time abnormal data judgment of the sliding window;

[0026] Among them, the specific operation steps of spatial dimension detection are as follows:

[0027] (a) For each data x i , respectively calculate x i The distance d(i,j) to other monitoring data in the cluster;

[0028] (b) Select an empirical value δ, count the number N of d(i,j)<δ;

[0029] (c) Compute with x i Adjacent node ratio P=N / N 0 , where N 0 is the total number of data in the cluster analyzed this time;

[0030] (d) Given the empirical critical value β, if P≤β, then determine the data x i for a...