Trustworthy Transmission And Channel Estimation For Relay And Intelligent Reflection Systems

Beyond 5th generation mobile networks(B5G)and 6th generation mobile networks(6G)provide high-quality communication services and flexible network structure.With the development of B5G network construction,network security cannot be ignored.With the enhancement of computational capacity,the traditional cryptography scheme implemented in the network layer is facing increasingly severe security challenges.This thesis focuses on "internal and external" security risks in communication system from the statistical properties of observations in physical layer,and proposes corresponding countermeasures.This thesis is divided into three parts.Firstly,this thesis studies the channel coding problems with the requirement of guaranteeing information integrity in the presence of malicious relays.In this thesis,we find a kind of "non-manipulable"channel condition in distributed relay system.Under this channel condition,the destination nodes can detect the attack caused by malicious relays or correct the error code caused by malicious relays in probability.Due to secure cryptographic key is not needed,the transmission efficiency is high.It is also proved that the binary symmetric channel(BSC)and binary erasure channels(BEC)can satisfy the " non-manipulable " condition.Finally,experimental results with Turbo code are carried out to verify the theoretical results.Secondly,focusing the attacks from external nodes on pilot and data phase transmission,this thesis studies the trustworthy estimation technology under these attacks.Specifically,this thesis first demonstrates the attack ability of the intelligent reflection surface(IRS),and analyzes the impact on channel estimation caused by these attacks.The IRS used by the attacker as an attack tool is called malicious IRS.Motivated by this result,this thesis utilizes the geometric arguments of the received signal to combat against these attacks from external nodes.The trustworthy channel estimation and information extraction are finally completed.The experimental results show that the proposed method has better performance than other methods.Finally,in order to further improve the estimation efficiency of intelligent reflection channel,this thesis focuses the malicious IRS networks,wherein the probability of random reflection could be shared before communications,and proposes a channel estimation method for multiple reflection channels.The proposed method allows multiple reflectors to be turned on at the same time,and uses channel separation technology to blindly separate all channels at the base station.Based on the channel separation,utilizing the reflection probabilities shared by nodes in network.Further an algorithm for completing channel identification based on the maximum probability criterion is proposed.In the end,identify the channel from each reflector to the base station.The proposed method does not need the channel identification information embedded in the reflected signal,and is suitable for the characteristics that the reflector device does not have the signal processing ability.The experimental results confirm the performance advantages of the proposed method.The proposed method also applies to collaborative communication of IRS.