Research On Secure And Energy Efficient Transmission Technologies For Internet Of Things

With the large-scale commercial deployment of the fifth generation(5G)mobile communication system and the launch of the research and development of the sixth generation(6G)mobile communication system,the application of the Internet of Things(IoT)has been booming,which is expected to realize the true Internet of Everything.At the same time,the security of the IoT is facing severe challenges.The information perceived by IoT devices is transmitted through wireless signals,which is vulnerable to eavesdropping,detection,and other attacks by illegal users.The physical layer security technology that utilizes the randomness properties of wireless channel is expected to provide wireless endogenous security that does not depend on high deciphering complexity for the future IoT.Therefore,how to design energy-efficient secure transmission mechanisms in energy-constrained IoT communication scenarios has become an urgent challenge to be solved.This dissertation focuses on the research on the energy-efficient and secure transmission technology of the IoT networks,including the research on the secrecy communication in the scenarios of cellular IoT and UAV-enabled IoT,and the research on the covert communication in the scenario of UAV-enabled IoT.Based on stochastic geometry,convex optimization and game theory,secrecy rate,screcy energy efficiency(SEE),and covert energy efficiency have been analyzed and optimized in energy-constrained IoT networks.Detailed contributions are summarized as follows:(1)To deal with the passive eavesdropping of simultaneous wireless information and power transfer(SWIPT)-enabled cellular IoT network,we propose a wireless resource allocation scheme based on multi-objective optimization(MOO).By jointly optimizing the uplink power allocation and the downlink beamforming vector,the SEE maximization and energy harvesting efficiency(EHE)maximization problem are formulated.Leveraging the weighted Tchebycheff method,the MOO problem between SEE and EHE is established.A two-layer resource allocation algorithm is proposed,in which the outer layer algorithm uses Dinkebach approach to solve the fractional programming problem,and the inner layer algorithm uses semi-definition relaxation(SDR)and successive convex approximation(SCA)to convert the non-convex problem into convex one.Simulation results show that SEE and EHE are conflicting design objective,and its optimal Pareto region increases with the number of antennas.Moreover,at the same downlink secrecy rate threshold,the proposed scheme achieves 16.5%SEE gain and 35.5%EHE gain compared with the directional beamforming scheme,and 38.4%SEE gain and 81.8%EHE gain compared with the half-duplex scheme.(2)To cope with the passive eavesdropping in the uplink communication scenario of the UAV-enabled IoT network,we propose a secure data collection scheme for UAV-assisted wireless sensor network(WSN).An adaptive secure transmission scheme based on Wyner coding is proposed when the instantaneous channel state information(CSI)of the eavesdropper is unknown.By jointly optimizing sensor scheduling,sensor transmit power,UAV transmit power,flight trajectory,codeword rate,and redundancy rate,a SEE maximization problem is established.By using probability theory,the analytical expressions of the secrecy outage probability and the connection outage probability are derived,and the closed-form expressions of the optimal codeword rate and optimal redundancy rate are obtained.The transformed problem is decomposed into four subproblems by using the block coordinate descent(BCD)technique,and these non-convex subproblems are coverted into convex ones by using Lagrange duality and SCA.The simulation results show that the proposed scheme can obtain almost the same secrecy rate gain as the sum secrecy rate maximization scheme under the lower UAV energy consumption.In addition,compared to the fixed trajectory scheme,the proposed scheme achieves significant SEE performance gain.(3)Focusing on the active eavesdropping in the downlink communication scenario of UAV-enabled IoT network,we propose a zero-sum differential game scheme between legal UAV and eavesdropping UAV.Dynamic differential equations for the UAV trajectories are given with characterizations for the positions,velocities,and acceleration.A zero-sum differential game is formulated to model the "pursuit-evasion" interaction between legal UAV and eavesdropping UAV,in which legal UAV aims to maximize the sum secrecy rate at the minimum power consumption cost while eavesdropping UAV aims to minimize the sum secrecy rate at the minimum power consumption.The definition and existence proof of Nash equilibrium(NE)of zero-sum differential game are provided.Pontryagin’s minimum principle is applied to solve the optimal trajectory control problem by introducing a Hamiltonian function and GaussSeidel-like implicit finite-difference method is leveraged to obtain the saddle point strategies at NE.The simulation results show that the proposed differential game reveals the secrecy performance limit of the UAV-assisted IoT system with active eavesdropping.Moreover,compared with passive eavesdropping with fixed strategy,the proposed scheme can achieve higher secrecy performance with lower communication-related energy consumption.(4)For the illegal detection in the relay communication scenario of UAVenabled IoT network,we introduce intelligent reflecting surface(IRS)to enhance covert transmission performance,and propose two covert data transmission schemes:UAV-Relay and UAV-IRS.Under the bounded noise uncertainty model,the miss detection probability,false alarm probability,and minimum detection error probability are analyzed and derived from the perspective of illegal detector.For the UAV-Relay transmission scheme,the covert transmission rate maximization problem is established by jointly optimizing sensor scheduling,UAV transmit power and flight trajectory,and an iterative optimization algorithm based on Penalty-SCA(P-SCA)is proposed.For the UAV-IRS transmission scheme,the covert energy efficiency maximization problem is formulated by jointly optimizing sensor scheduling,sensor transmit power,IRS phase shifts,and flight trajectory,and an iterative optimization algorithm based on alternating optimization(AO)is proposed.The simulation results show that the deployment of dynamic UAV can effectively improve the covertness performance,and the flight trajectory of UAV is alway far away from the detector.Furthermore,when the number of IRS reflection elements is large,the UAVIRS scheme has a larger covert energy efficiency performance advantage than the UAV-Relay scheme.