Identity authentication method, storage medium and device integrating user's multi-source voice characteristics

Abstract

The invention discloses an identity authentication method, storage medium and equipment that integrates the user's multi-source voice characteristics, and constructs a sensing end; designs a sending and receiving protection mechanism, and adds it when the signal is transmitted during the tag communication process between an RFID reader-writer and the sensing end. Dynamic random noise, eliminate the noise and restore the label EPC when receiving the signal, and judge the authenticity of the perceived signal; recognize the user's voice command, if it contains keywords, determine whether to enter the next step or reject the request and end according to whether the sensory end label EPC can be restored correctly ; Perform noise reduction and vocal part detection on the voice signal, convert it into a time-spectrogram and extract the phase information of the RF signal and preprocess it; input the processed voice and RF sensing data into the neural network through two branches at the same time for user authentication . The realization of the present invention based on voice and radio frequency technology can not only ensure the safety and reliability of the system, but also has strong application value, social significance and development potential.

Description

technical field [0001] The invention belongs to the technical field of radio frequency identification (RFID) and voice authentication, and in particular relates to an identity authentication method, a storage medium and a device that integrate user multi-source voice characteristics. Background technique [0002] In real life, most access control systems use radio frequency identification cards to authenticate users. However, as radio frequency identification is a non-user-binding authentication method, the card itself and the radio frequency signal face security risks such as loss, theft, tampering and copying (replay). In recent years, fingerprint, face and voice authentication technologies have been developed and matured, and have been adopted by more access control systems relying on better convenience and security. [0003] However, in the face of some special application scenarios, especially contagious public health events, fingerprint authentication will directly le...

AI Technical Summary

Problems solved by technology

[0003] However, in the face of some special application scenarios, especially contagious public health events, fingerprint authentication will directly lead to the risk of infection; face authentication will affect the accuracy of authentication due to the user wearing a mask to cover the face, and there is still a problem of privacy leakage; although voice authentication can Avoid the above-mentioned similar problems, but when the user speaks, the voice command is easily recorded by the attacker and initiates a replay attack, and the cost of implementing a voice replay attack is low, and does not require highly specialized equipment and technical requirements

Images