Research On Resource Optimization For Full Duplex Ambient Backscatter Communication

As Internet of Things(IoT)becomes more pervasive,some IoT devices face high energy consumption and battery power issues.As a low-power green communication technology,Ambient Backscatter Communications(AmBC)have received much attention.Backscatter Device(BD)doesn’t need to generate radio frequency signal but can use the existing RF signal in the environment to transmit information using reflection modulation.BDs can be used as low-power sensing devices in the IoT.Most of the existing studies on AmBC focus on a single antenna system or a single BD scenario,while there are relatively few studies on the optimal allocation of communication network resources in multi-antenna and multi-BD scenarios.Therefore,this thesis investigates resource allocation optimization algorithms for ambient backscatter communication systems with multiple antennas and full duplex transmitters.This thesis is dedicated to research on resource allocation problems for three different multi-antenna and multi-BD Amb Cs.First,this thesis aims to maximize the minimum throughput among all BDs in an AmBC containing a multi-antenna full-duplex access point,a downlink user,and multiple BDs.Self-interference residuals at the access point and the minimum downlink user rate constraint are considered.Then,an iterative algorithm is proposed by jointly optimizing beamforming,transmission time,and reflection coefficient.Second,a robust resource allocation algorithm is proposed for maximizing the minimum throughput among all BDs under a bounded channel uncertainty model to address the channel estimation error issues.Considering the channel estimation error of two-path cascade channels and assuming that the error is within the bounded elliptic closed set,the robust optimization problem containing uncertain constraints is reformulated into a tractable problem based on the worst-case principle and S-procedure.Beamforming and reflection coefficient are alternately optimized based on the block coordinate descent method.Semi-positive definite relaxation and continuous convex approximation methods are used to solve the subproblem of beamforming and derive the analytical formulas for the optimal reflection coefficient and time assignment.Third,to tackle the challenges of high energy consumption and data eavesdropping in wireless communication systems,a secure backscatter communication model is proposed,the model consists of a full-duplex base station,a downlink user,an eavesdropper,and multiple-BDs.Meanwhile,the minimum rate constraint of the downlink user,the minimum secure rate constraint of BDs,the harvesting energy constraint,and the maximum signal-to-interference-plus-noise ratio constraint of the eavesdropper are considered.Then,a resource allocation algorithm based on Charnes-Cooper transformation,Dinkelbach transformation,semidefinite relaxation,and successive convex approximation is proposed.The performance difference between fixed zero-forcing beamforming and optimized beamforming algorithm is also investigated.Numerical results demonstrate the effectiveness of the proposed algorithms.