Physical Layer Key Extraction Based On Sensing Signals

With the wide deployment of Internet of Things devices in wireless communication system,higher security requirements are put forward by users.On the one hand,the openness of wireless channel brings potential risks such as data leakage.On the other hand,through the utilization of wireless channel features,Io T devices can realize data encryption and identity authentication based on physical layer key extraction technology.The technology not only meets the requirements of the rapid development of social production and life,but also ensures the security of wireless communication system.However,the existing physical layer key extraction technology still has some inherent problems such as bit generation rate is low,bit randomness is poor and bit mismatch rate is high.In addition,each wireless service is associated with a certain region of the radio spectrum,and the technology of physical layer key extraction that depends on the radio channel is also associated with the radio spectrum.However,the application of wireless communication technology is increasing unprecedentedly,which leads to the serious spectrum congestion problem.In order to alleviate the above problems,this thesis proposes a physical layer key extraction scheme based on sound signal and visible signal.The main research contents are listed as follows:1.Due to the design and limited optimization of quantization algorithm and information harmonic algorithm,the key extraction efficiency is low in the existing schemes based on sound signal.In this thesis,an efficient key extraction scheme based on sound signal is proposed by adopts A law 13 broken line quantization algorithm optimized by window function and the information harmonic algorithm of polarization code error correction.In this scheme,the quantization algorithm improves the bit generation rate and bit randomness,and the information harmonic algorithm reduces the bit mismatch rate.In addition,the effectiveness of our scheme is verified by using Internet of Things devices.Experimental results show that the key generated by the new scheme can passes the NIST randomness test.Compared with similar related schemes,the bit generation rate of the new scheme increases about 10% and bit mismatch rate reduces about 80%,thus realizing efficient key extraction.2.Although the existing key extraction schemes can improve the efficiency of key extraction by optimizing the algorithm,they cannot alleviate the problem of spectrum congestion.In this thesis,we propose a visible light key(VLC)extraction scheme based on adaptive quantized.The scheme adopts the existing basic lighting facilities to transfer the communication spectrum to the visible spectrum,which effectively alleviates the problem of spectrum congestion and ensures the efficiency of key extraction.The adaptive quantization of the scheme combines the visible light signal with the radio frequency signal to increase the bit randomness,and adopts polarization code to strengthen the main communication channel,which degrade the communication quality of the eavesdropping channel and improves the efficiency of key extraction.At the same time,the communication efficiency of visible light in indoor environment is analyzed through simulation experiments and the performance index comparison between the scheme and relevant key extraction scheme is listed.The analysis results show that the bit error rate of the scheme is 0.0184,and the bit generation rate is about 340 (74)/.Which realized high efficiency key extraction.