LED spectrum fingerprint feature individual identification method and device based on linear attention network

Through the LED spectral fingerprint feature individual recognition method based on the linear attention network, the problems of counterfeiting and tampering of pseudo base stations in the visible light communication system are solved, and physical layer security authentication, indoor positioning and fixed asset management are realized.

CN119995946BActive Publication Date: 2025-10-17Chinese People's Liberation Army Cyberspace Force Information Engineering University
2 Cites 0 Cited by

Patent Information

Application Number
CN202510045442.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-01-10
Publication Date
2025-10-17
Estimated Expiration
2045-01-10

AI Technical Summary

Technical Problem

In existing visible light communication systems, unauthenticated LED lights (fake base stations) pose security risks of counterfeiting and tampering, which cannot be effectively addressed by traditional communication protocol authentication methods.

Method used

An LED spectral fingerprint feature individual recognition method based on a linear attention network is adopted. By collecting, preprocessing and training the spectral data of LED light sources, the SpecLinNet network is used to extract feature vectors and perform cosine similarity comparison to achieve individual recognition and authentication of LED light sources.

Benefits of technology

It achieves physical layer security authentication, and the LED spectrum fingerprint characteristics cannot be tampered with or counterfeited, which improves the security of the visible light communication system and supports indoor positioning and fixed asset management.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present application belongs to the field of visible light communication, and relates to an LED spectrum fingerprint feature individual identification method and device based on a linear attention network. The method comprises the following steps: 1) collecting original spectrum data of an LED light source and forming a spectrum fingerprint database; 2) pre-processing the original spectrum data of the LED light source, constructing a training data set, training using a SpecLinNet network, inputting the training data set into the SpecLinNet network which has been trained, extracting a feature vector before a Softmax layer, and forming a feature vector library; and 3) firstly performing weighted multi-element scattering correction pre-processing on original spectrum data of an LED to be identified which is obtained from a spectrum collector, inputting the data into the SpecLinNet network, comparing the feature vector output by the network with features in the feature vector library one by one in terms of cosine similarity, and performing judgment. The method is a physical layer security authentication method and has higher security.
Need to check novelty before this filing date? Find Prior Art