Distributed intrusion detection method based on multi-layer extreme learning machine in Internet of Things environment

Abstract

The invention discloses a distributed intrusion detection method based on a multi-layer extreme learning machine in the environment of the Internet of Things. Due to the resource-constrained characteristics of related equipment, such computationally heavy tasks for realizing automatic attack detection require Move to edge devices to keep processing close to data sources. These edge devices can run pre-built classification models. But when faced with a large amount of training data, there is not enough storage and processing power to construct and upgrade such models. In order to solve this problem, the present invention moves the computation-intensive and large-storage training calculations to the cloud server, constructs and trains single-hidden-layer extreme learning machine and multi-hidden-layer extreme learning machine models in the cloud server, so that the boundary The device performs traffic classification based on the deep learning model preset in the cloud server, so as to classify whether it is normal traffic or network attack, and through experimental analysis, it is concluded that the multi-hidden layer extreme learning machine has better performance.

Description

technical field [0001] The invention belongs to the field of intrusion detection and deep learning, and in particular relates to a distributed intrusion detection method based on a multi-layer extreme learning machine under the Internet of Things environment. Background technique [0002] The rapid development of Internet of Things technology is rapidly bridging the gap between traditional information services and the surrounding physical environment through the connection of increasingly complex sensing devices and Internet-based remote control devices. Many potential IoT application services have emerged, such as environmental monitoring, traffic monitoring, health care monitoring, etc. These applications have greatly improved the interaction between humans and computing devices. The growing Internet of Things applications and physical information services have increasingly high requirements for network security. Intrusion detection under the Internet of Things has become ...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problem in the prior art that the border devices of distributed intrusion detection in the Internet of Things environment cannot handle too large training data, the present invention provides a distributed intrusion detection method based on multi-layer extreme learning machines in the Internet of Things environment

Images