Method for detecting failure nodes of wireless sensor network

Abstract

The invention provides a mechanism for detecting failure nodes of a wireless sensor network based on high reliability, which is used for detecting the failure nodes in the wireless sensor network and finding out the types of the failures, wherein the failures include the Byzantine failure, namely uncertain failure with various reasons. The mechanism is characterized by sending a detection data packet to nodes to be detected, using a threshold-based failure node detection algorithm for the nodes to be detected, reporting a failure and requiring further processing of the failure after the failure is detected, and continuing detecting the next node if no failure is detected, thus ensuring that all the failures are detected, wherein the detection data packet is processed by data mining and data fusion technologies. Meanwhile, a failure detection priority is set so as to ensure that the detection priority can be given to the fatal failures.

Description

technical field [0001] The invention relates to the detection, perception and notification of faulty nodes during multi-node interactive communication in a wireless sensor network, so that customers can find faulty nodes in time when nodes cannot communicate interactively or obtain inaccurate test results, and improve communication efficiency and communication quality, this technology belongs to the cross-technical application field of computer network, artificial intelligence and wireless sensor network. Background technique [0002] The wireless sensor network is composed of a large number of low-cost miniature sensor nodes deployed in the monitoring area. A multi-hop mobility peer-to-peer network is dynamically formed by several nodes using wireless communication, so that it does not depend on any infrastructure and has distributed, Features such as self-organization, topology change, multi-hop routing, and poor security are aimed at cooperatively sensing, collecting, and...

AI Technical Summary

Problems solved by technology

[0016] (1) The energy of the battery is limited. Due to the miniaturization of the node, the energy of the battery of the node is limited, so that most of the normal nodes cannot work normally due to energy depletion, resulting in failure

[0017] (2) The node is arranged in a harsh environment, and the probability of node failure increases due to changes in the environment

[0018] (3) With the development of wireless sensor networks, the accuracy and efficiency of measurement results are increasingly sought after, and a large number of faulty nodes will affect the feedback of results

[0019] (4) Faulty nodes will generate wrong data, which will make the monitoring center get wrong results and affect its safety. For example, when measuring volcanic eruptions, if the faulty node gives wrong feedback, it will bring certain security risks

The faulty node inspection method based on fuzzy theory is a method of detection according to the human thinking process. It is mainly used to measure the problem of less data and cannot obtain an accurate model. It is a solution to deal with uncertainty, but its establishment Correct fuzzy rules and membership functions are very difficult

Images