Method and system for weak electromagnetic interference attack detection of Internet of Things based on edge computing

Abstract

The invention provides a method and system for detecting weak electromagnetic interference attacks of the Industrial Internet of Things based on edge computing, belonging to the field of the Industrial Internet of Things. The method includes: extracting the noise fingerprint of the smart device from the sensing data transmitted from the sensing module to the MCU as a sample, and extracting M pieces of noise with a length T under two states of no weak electromagnetic interference attack and weak electromagnetic interference attack Fingerprint to establish a noise fingerprint sample library, where M≥5000; establish a FEU-LSTM fusion model, the FEU-LSTM fusion model includes a convolutional neural network-based FEU model and a time-recurrent neural network LSTM model; randomly extract from the noise fingerprint sample library P noise fingerprints are input to the FEU-LSTM fusion model for Q times of Epoch iterations to complete the model training, where P≥125, Q≥30; during the attack detection process, the length of the real-time read from the sensing data is T Input the noise fingerprint of the device into the trained FEU-LSTM fusion model, and judge whether the smart device is attacked by weak electromagnetic interference according to the output result of the model.

Description

technical field [0001] The invention relates to the field of the industrial Internet of things, in particular to a weak electromagnetic attack detection method of the industrial Internet of things and a weak electromagnetic attack detection system of the industrial Internet of things. Background technique [0002] The current intelligent industrial production is inseparable from the support of the Industrial Internet of Things. The Industrial Internet of Things has very high requirements on the reliability and security of data. Once the data in the Industrial Internet of Things system is tampered with, it will affect the quality control of products. even cause major accidents. [0003] In a smart device, the process from data collection to transmission includes: the sensing module converts physical quantities into analog electrical signals, and then sends the signals to the sampling module MCU module through a signal line, and the MCU module receives the data and performs se...

AI Technical Summary

Problems solved by technology

[0005] In industrial sites, attackers can use specific electromagnetic interference devices to change the voltage on signal lines to achieve the purpose of quietly tampering with data. This kind of attack can easily destroy the accuracy of industrial equipment and lead to production accidents

Since the devices in the Industrial Internet of Things (such as analog-to-digital converters) are easy to couple with electromagnetic waves, and the signal lines for communication between the sensing module and the MCU act as antennas, attackers can launch weak electromagnetic interference attacks in various ways, so weak electromagnetic interference Interference can tamper with the data generated by the perception module

[0006] In order to defend against electromagnetic interference, electromagnetic interference filters and shielded wires are currently used to defend against the impact of strong electromagnetic interference. Although this type of electromagnetic interference defense strategy can reduce the impact of strong electromagnetic interference, electromagnetic waves will still penetrate the circuit system. That is, the defense strategy cannot eliminate the negative effects of weak electromagnetic waves

However, the lack of data security verification and attack detection between the sensing module and MCU makes it impossible to detect attacks caused by weak electromagnetic interference, which poses a greater risk to the safety and reliability of industrial production

Images