A Trust Model Based on Isolation Forest in UASNS

Abstract

The invention belongs to the technical field of wireless sensor network security, and discloses a trust model based on an isolated forest in UASNs. First, the trust items to be evaluated are divided into four types: communication trust, data trust, energy trust and environment trust. In particular, in the process of data trust calculation, in addition to the difference between a single node and the overall data distribution, the The similarity between the overall data distribution and the theoretical data distribution in UASNs. Since environmental noise will affect the communication and data transmission between nodes, which will cause energy changes, increasing environmental trust can reduce the evaluation error of other trust items and improve the accuracy of malicious detection; then, the four trust items are used as trust training sets, The isolated forest algorithm is used to establish a node trust model; finally, based on the node trust model, the reliability of nodes in UASNs is evaluated to effectively detect abnormal nodes and malicious node attacks in UASNs and reduce network security threats.

Description

technical field [0001] The invention belongs to the technical field of wireless sensor network security, in particular to an isolated forest-based trust model in underwater acoustic sensor networks (UASNs). Background technique [0002] Underwater Acoustic Sensor Networks (UASNs) can play meaningful roles in various fields, such as battlefield surveillance, smart ocean, emergency response, target tracking, and marine ranching, etc. UASNs is a multi-hop self-organizing distributed sensor network, which is composed of a large number of tiny sensor nodes with sensing capabilities. Usually, sensor nodes are randomly deployed in unattended areas and harsh environments to perform various complex tasks and are vulnerable to Different kinds of security threats. Although traditional security methods based on encryption or authentication can defend against external attacks, they cannot effectively detect internal attacks from compromised nodes. For example, in some cases, by capturi...

AI Technical Summary

Problems solved by technology

For example, in multidimensional trust, researchers are required to assign impact factors subjectively, which leads to inaccurate trust evaluation; some studies only combine direct observation and indirect observation to calculate trust without considering energy constraints, which may shorten the life of the network; In addition, although some studies have considered the movement of node positions caused by water flow movement, none of the existing work has considered the impact of environmental noise (such as turbulence, boat activities, wind and thermal air flow) on trust evaluation calculations.

Images