A method and system for securely accessing IoT devices based on RFID signals

Abstract

The present invention relates to a secure access method for IoT devices based on RFID signals, comprising: collecting changes in RFID signals caused by users taking actions in the pattern database; extracting corresponding RFID features to construct an action recognition training set; constructing And train the CNN-LightGBM model; after receiving the RFID signal sent by the IoT device to be connected, extract the corresponding RFID feature; input the extracted RFID feature into the CNN-LightGBM model; the CNN-LightGBM model outputs the action category and prediction Probability; judge whether the output action category and the time of receiving the RFID signal sent by the IoT device to be connected match the action in the behavior pattern database and the corresponding time period, and if so, agree to the access request of the IoT device to be connected; otherwise reject the pending connection Access requests from IoT devices.

Description

technical field [0001] The present invention relates to the technical field of the Internet of Things, and more specifically, to a secure access method and system for IoT devices based on RFID signals. Background technique [0002] With the continuous development of Internet of Things (IOT) technology, more and more IoT devices appear in people's lives. Most of these devices are located in the user's home, such as smart air conditioners, smart door locks, smart speakers and so on. These IoT devices together constitute the user's smart home environment, providing users with an unprecedented experience in their home life. However, a large amount of sensitive data generated during the operation of these IoT devices will make user privacy more likely to be leaked. Therefore, ensuring the safe operation of these IoT devices is an important capability that must be deployed for each device. The most important point is that newly deployed IoT devices must be able to establish con...

AI Technical Summary

Problems solved by technology

The first problem is that there is a lot of noise that is difficult to eliminate in the method based on the Wi-Fi signal. The original CSI information of the Wi-Fi signal collected by the existing technology will have a lot of noise, such as cyclic shift diversity, sampling time offset, Noise caused by sampling frequency offset and beamforming, etc., these noises are difficult to remove, making it difficult to extract signal fluctuation features with sufficient confidence

The second problem is that Wi-Fi itself is vulnerable to attacks, and accessing Wi-Fi to maintain a long-term connection violates the security principle of the security domain. The existing method requires IoT devices to be connected to the Wi-Fi LAN in the security domain first. Only by establishing a Wi-Fi connection can Wi-Fi signal fluctuations be obtained, which allows unauthenticated devices to freely access the LAN in the security domain and exchange information with the security domain, which seriously violates the security principles of the security domain

The third problem is that the existing method based on the extraction of feature strings takes a long time to access and is uncontrollable. The existing method continuously detects the changes in the radio frequency signal caused by the movement in the room, and extracts the signal features to generate a symmetrical key

However, due to the multipath effect, the multipath of the RF signals transmitted by IoT devices at different locations is very different; even though they are located in the same room, it is very difficult to generate a symmetric key by extracting a characteristic string

If the characteristic strings are required to be exactly the same to generate a symmetric key, the generation of a symmetric key can only be achieved by fluctuating through a similar multipath radio frequency signal affected by an action at a specific location, which will take a long time, and Uncontrollable length of time spent

Images