Adaptive Hierarchical Intrusion Detection Method for Dynamic Topology-Oriented Wireless Sensor Networks

Abstract

The invention discloses an adaptive layered intrusion detection method for a dynamic topology-oriented wireless sensor network. Firstly, a wireless sensor communication network is established. When a target node is connected to the network, the underlying common node is used as a discriminant node to perform an intrusion based on a single spatiotemporal correlation. The detection mechanism uses the temporal-spatial correlation of data packet characteristics to conduct preliminary legality identification; if it is judged to be suspected to be legal, the cached data and subsequent data are allowed to be forwarded upwards, and the data is marked; if it is judged to be suspected to be illegal, the T1 time The feature results in the segment are sent to the aggregation base station, which makes the final judgment; the center aggregation base station executes the center identification algorithm based on the multi-temporal and spatial correlation of the isolated forest, makes a comprehensive decision, and realizes the final judgment on the legitimacy of the target node; and The final judgment is fed back to the underlying discrimination node, and the discrimination node realizes the access and isolation of the target node. This method can realize network adaptive intrusion detection with light energy consumption.

Description

technical field [0001] The invention belongs to the technical field of sensor network security, in particular to a dynamic topology-oriented network self-adaptive intrusion detection method. Background technique [0002] After the base station is deployed in the wireless sensor network, its sensing data has the characteristic of converging to the base station. Affected by external factors, the nodes in the wireless sensor network may leave the network after being forced to move, and then need to re-connect to the network, which makes the network have the characteristics of dynamic topology evolution. Illegal nodes will also take the opportunity to access the network, steal data, and destroy the authenticity and reliability of data sources. At the same time, as an important support for multiple information neighborhoods, wireless sensor networks, especially in special fields such as military early warning, need to have the ability to prevent illegal node intrusion, ensure th...

AI Technical Summary

Problems solved by technology

The advantage is that the detection quality is high, and the disadvantage is that the computing resources are too large

[0004] Considering the limitations of some distributed isomorphic wireless sensor networks and the characteristics of simple network data packets and few data features, it is difficult for a single anomaly detection method and structure to balance low resource load and high detection efficiency at the same time. Accuracy

Images