Key Generation Based On Wireless Channel Characteristics

Encryption is the main tool for implementing communication security.At present,in addition to quantum communication,the traditional communication encryption system can only achieve computational security.With the development of quantum computers,this encryption system is facing a huge test.Meanwhile,the generation,management and distribution of secret key not only costs a lot of energy of the user and administrator,but also hidden dangers that artificially leaked key.In addition to openness,the terrestrial wireless channel also has the characteristics of reciprocity,temporal variability and spatial variability.Based on the above-mentioned wireless channel characteristics,the both users of wireless communication can generate two identical random number sequences through channel parameter acquisition,quantization,negotiation and so on.The sequence can be used as a key seed for traditional encryption schemes after processing.The entire process of key generation and distribution does' t need manual participation,reducing the risk of key leakage and the burden of key management.Importantly,if the method combined with the system of one-time pad,it is possible to achieve the so-called information-theoretic security that is unconditional security.This paper studies and analyzes the theoretical problems,the quantization algorithm of channel parameters and the key negotiation algorithm of the key generation based on the wireless channel.The main work is as follows:1.We analyzed the feasibility of generating the key based on the wireless channel by using the information theory.It is shown that if the third party can only launch the passive attack,the two legitimate parties can generate secret key,but if the third party can initiate the active attack,the possibility of a wireless channel to generate secret key depends on the presence or absence of the simulatability condition.In addition,since the shared key is generated by the open channel,in order to ensure the privacy of the key,the privacy enhancement protocol has been studied.The research shows that the private enhancement protocol based on the extractor can effectively improve the secret characteristic of the key.2.Aiming at the problem of high complexity and threshold offset asymmetry in CQA(Chanel Quantization Alternating)algorithm,a single threshold mobile CQA algorithm is proposed by analyzing the conditions of threshold movement.According to the simulation results,the improved algorithm can obtain the same quantization performance as the original algorithm when the channel resource is saved by 25%.In addition,an improved quantization algorithm based on the log likelihood ratio of joint probability is proposed to solve the problem that the high bit error rate affects the key generation length in CQA.It can be seen from the simulation that the key length in theory of quantization algorithm based on the log likelihood ratio of joint probability increases by at least 0.1bits / symbol compared to the original algorithm when the channel resource is saved by 25% and the correlation coefficient is less than 0.7.3.In order to solve the problem that the quantized hard decision sequence as the decoder input affects error correction performance in the key negotiation,a negotiation algorithm based on the CQA is used to calculate the quantized soft information.The algorithm calculates the the quantized soft information of each bit by using the position index of sender and the sample value of receiver,and takes the quantized soft information as input to the Belief Propagation(BP)decoder of low density parity check(LDPC)code.The error correction capability based on the quantized soft information negotiation algorithm is improved by 33% compare to the negotiation algorithm based on the quantified hard decision sequence.4.In order to solve the problem that LDPC code has a lower error correction capability when the codeword length is short,a Polar code negotiation algorithm based on successive cancellation list(SCL)decoding is proposed.It can be seen from the analysis that the error correction capability based on Polar negotiation algorithm is improved by 21% compare to LDPC code when the codeword length is short.In addition,the channel coding may be produces the error diffusion when the initial error is relatively high,so the negotiation algorithm that Binary combined Polar code is proposed.The algorithm uses the characteristic that Binary won't produce the error diffusion when the initial error rate is high.First,Binary is used to correct the error to reduce the initial error rate,and then use the Polar code to negotiate the key.The simulation results show that the error correction capability of the joint algorithm is improved by 24% compare to Polar code.However,because the joint algorithm enhances the error correction capability by sacrificing the key length,therefore,in the consultation should be a reasonable choice of negotiation algorithm.