| With the large-scale deployment of 5G mobile communication networks,the contradiction between the surging number of wireless devices and the energy consumption of communication infrastructures has become increasingly prominent.The traditional technologies aiming at transceivers are gradually entering a bottleneck to tackle this contradiction.In recent years,channel enhancement technology represented by intelligent reflecting surface(IRS)has become a research hotspot.It realizes constructive multipath signals and enhances the equivalent channel between transceivers by adjusting the phase shifters of a large number of integrated reflecting units.Furthermore,the IRS empowered by wireless power transfer technology can not only significantly alleviate the energy consumptions of the communication systems,but also requires no dedicated energy supply facilities,which can be regarded as a systematic solution to tackle the energy consumption problem of the next generation mobile communication systems.In this thesis,the wirelessly charged IRS is used as one of the infrastructures of the future mobile communication systems,and a prospective theoretical study is carried out on the overall energy efficiency(EE)of this new type of mobile communication system.The main research challenges include:(1)Selection of wireless charging schemes(i.e.,time-switching and power-splitting)for charging the IRS.(2)With given charging scheme,the problem of balancing charging and assisting transmission,multi-antenna transmitter beamforming design and IRS phase control to maximize energy efficiency.(3)The robustness of the system after the ideal communication channel is extended to a channel with uncertainties.To tackle these challenges,this thesis conducts a comparative study via two technical routes and extend the EE optimization problems under ideal channel conditions to the robust optimization problems under channel estimation errors.The main contributions and innovations of this thesis include:(1)For the EE optimization problem under the time-switching scheme,a joint optimization algorithm of time slot allocation,transmitter beamforming and IRS phase shift is proposed,which achieves the maximum throughput for a given transmission power.Specifically,for the non-convex optimization problem with coupled variables,a twostage algorithm is proposed,which decouples parts of variables.Combined with the monotonic optimization framework,an iterative algorithm is proposed to solve the problem efficiently.(2)For EE optimization problem under the power-splitting scheme,a joint optimization algorithm of amplitude reflection coefficient,transmitter beamforming and IRS phase shift is proposed,which minimizes the transmitter power and subjects to signal-tonoise ratio requirements at the receivers.To tackle the complex non-convexity,a twolayer iterative algorithm based on the alternating optimization framework is proposed,and the update strategy of the amplitude reflection coefficient is devised to realize the fast convergence of the algorithm.(3)For the above two technical routes under the ideal channel assumptions,a robust optimization model under the bounded channel estimation error is proposed to ensure the robustness of the system performance.The influence of the channel estimation error on the system performance is quantitatively evaluated via simulation results and the necessity of robust optimization is elaborated.Through the comprehensive and comparative study of the above two technical routes,this thesis reveals that: When the IRS energy consumption constraints are relatively easy to meet,the power-splitting scheme can provide better EE performance for the IRS-assisted communication systems.While the IRS energy consumption constraints are more stringent,the time-switching scheme can obtain a more stable EE performance.The above research findings can serve as important reference for practical deployments in the near future. |
PDF Full Text Request