Beamforming Design And Optimization In Intelligent Reflecting Surface-assisted SWIPT System

Intelligent Reflecting Surface(IRS)is a planar array composed of multiple reconstructed passive reflective elements,whose working mode is coordinated by a software controller.Each element in IRS controls the reflection angle and reflection intensity of incident electromagnetic waves.Hence,the phase and amplitude of the reflected signal can be correspondingly changed.The reflected electromagnetic wave can form a three-dimensional passive beam with independently phase-shift,which meets the requirements of different communication scenarios.IRS acts as an intelligent passive relay,and it provides low power consumption and high energy-efficiency communication paradigm for the future 6G wireless communication network.It is considered as one of the most promising wireless communication technologies.Simultaneous Wireless Information and Power Transfer(SWIPT),which allows information and power to be transmitted simultaneously,has brought the transformation of energy-constrained communication networks.Joint design and optimization of energy-efficiency priority based active beamforming at multi-antenna transmitter and passive beamforming at IRS are investigated in Chapter 2.The optimization function is to maximize energy-efficiency at Information Decoding(ID)receiver,which is subject to transmitter power control,energy constraint at EH receiver and IRS phase shift constraint.Alternating Direction Method of Multipliers(ADMM)is implemented to solve this non-linear non-convex optimization problem with multi-variable coupling.Firstly,IRS phase shift matrix is fixed and the fraction objective function is transformed into the form of subtractive parameters by using Dinkelbach’s approach.Singular Value Decomposition(SVD)and Semi-definite Relaxation(SDR)are implemented to obtain active beamforming vector at transmitter.Once the optimal active beamforming vector is obtained,auxiliary vector is introduced to transform energy-efficiency optimization problem into convex optimization,and IRS phase shift matrix as well as reflective beamforming vector can be derived via SDR method.Simulation results show that,IRS assisted SWIPT could significantly reduce the energy threshold at EH receiver.Maximum energy-efficiency can be achieved with the maximum numbers of passive reflecting elements and transmitter antennas.Cascaded channel estimation approach combined with active beamforming at transmitter and passive beamforming at IRS is studied in Chapter 3.Firstly,cascaded channel estimation is transformed into the sparse signal reconstruction problem.Then,Compressed Sensing(CS)signal reconstruction approach is implemented to solve this problem.Only a small amount of training overhead provides reliable channel estimation gains as well as better beamforming gains.A second order rectified Parallel Factor(PARAFA)model is applied into IRS-aided SWIPT MIMO system,which is described by tensor decomposition method.The received signal can be represented via PARAFA model through algebraic structure form in time-domain and IRS phase shift.Two cascaded channel estimation approaches,namely,Bilinear Alternating Least Squares(BALS)and Least Squares Khatri-Rao Factorization(LSKRF),are proposed respectively.Simulation results show that,compared with Orthogonal Matching Pursuit(OMP)approach,the Normalized Mean Square Error(NMSE)of the proposed BALS cascaded channel estimation approach achieves the minimum value,which obtains better channel estimation and convergence performance with the least parameter constraints.IRS-assisted SWIPT model over Rayleigh fading channel with imperfect CSI is investigated in Chapter 4,which mainly focuses on the optimization of IRS reflection array spatial correlation and beamforming.Meanwhile,channel hardening characteristics are also analyzed in IRS-assisted SWIPT system.Simulation results reveal that,the enhancement of Signal-to-Noise Ratio(SNR)is proportional to the improvement of IRS reflection elements.In the direct path,the effect of IRS phase optimization is more significant,so that the communication quality is improved.In addition,the dissertation analyzes the channel hardening problem of IRS-assisted SWIPT with direct path and indirect path,and the improvement of communication performance.The dissertation conducts a preliminary study on the design and optimization of beamforming and channel estimation in IRS-assisted SWIPT.The research achievements provide reference for further related work.