Distributed detection data error attack defense method based on trust evaluation mechanism

Abstract

The invention discloses a distributed detection data error attack defense method based on a trust evaluation mechanism. An existing defense method is limited by many conditions, and the convergence and the convergence precision of a distributed system cannot be ensured. According to the node state updating rule and the neighbor node information in the distributed system, the state constraint rangecan be obtained. The credibility of the neighbor node is evaluated by observing whether the state of the neighbor node is within a constraint range. The method comprises the following steps: sensinga target to obtain measurement data, detecting the abnormal behavior of a neighbor node, evaluating the trust degree of the neighbor node, identifying the identity of the neighbor node, adjusting thefusion weight, executing a consistency algorithm to update the node state and add compensation data. The credibility is evaluated only by utilizing data provided by local and single-hop neighbors, sothat malicious nodes are identified and isolated, and error data injected by the malicious nodes are offset through a recovery mechanism. The method has the advantages of being low in calculation complexity, free of network global information, capable of guaranteeing convergence precision and the like.

Description

technical field [0001] The invention belongs to the field of distributed information processing, in particular to the field of distributed detection in distributed information processing, and relates to a distributed detection data error attack defense method based on a trust evaluation mechanism. Background technique [0002] Distributed detection is one of the important applications of distributed information processing. Distributed detection means that the sensor nodes in the distributed network complete the detection task in a cooperative manner: the nodes first perceive the target to obtain measurement data, then exchange data with each other and perform data fusion through a consensus protocol, and finally judge whether the target is in the area of ​​interest . Distributed detection improves the detection accuracy and detection efficiency of the system as a whole. [0003] However, due to the open nature of distributed systems, sensor nodes are vulnerable to attacks ...

AI Technical Summary

Problems solved by technology

[0005] The above defense methods face many limitations in practical application: the local information of nodes is limited, and it is neither realistic nor safe for each node to obtain global network information; node resources are limited, and each node’s computing power, storage space, battery power, etc. Resources are limited; the network topology is uncertain, and the actual network topology may not meet the requirements of the above detection methods

Moreover, the above defense methods only detect whether there is an attacker or identify malicious nodes, but the wrong data that was injected still remains, resulting in deviations between the convergence results and the real convergence results

Images