Beamforming Design For Multi-Antenna Wireless Power Transfer Systems

With the evolution of the fifth generation(5G)networks and the development of Internet of things(IoT),the number of low-cost devices is growing explosively.Compared with the limited charging based on battery or connecting ables,the wireless power transfer(WPT)via radio-frequency(RF)signals has emerged as a promising solution to provide convenient energy supply.With the advantages of flexible,longer transfer distance,supportable for multiple devices,and applicable in non-line of sight environment,WPT has been widely adopted in wireless communication networks.However,due to the severe path loss of radio signals propagation and the nonlinearity of the rectifier,the energy transmission efficiency is highly limited.By deploying multiple antennas at the energy transmitter(ET)and designing the transmit energy beamforming,the transmission efficiency in wireless channels can be improved.Accordingly,how to combine the WPT technique and multiple antennas technique is meaningful for both the WPT and wireless communication networks.We focus on the high energy transfer efficiency under the non-linear energy harversting(EH)model and the application of WPT in wireless communication network.Accordingly,this thesis investigates the multi-antenna energy receiver(ER)architecture,the joint energy beamforming and power splitting optimization in point-to-piont multiple-input multiple-output(MIMO)WPT system,the fairness problem in multiuser multiple-input single-output(MISO)WPT system and the wireless channel estimation,as well as the wireless powered over-the-air computation system.The main contributions of this thesis are as follows.(1)Due to the non-linear EH model,the conventional ER architectures cannot efficiently exploit the RF energy from the ET.To tackle this issue,we propose a novel generic ER architecture,in which one power splitter is inserted after each antenna to adaptively split the received RF signals towards the rectifiers.Then,we consider the point to point MIMO WPT system and aim to maximize the harvested direct current(DC)power by jointly optimizing the transmit energy beamforming at the ET and the power splitting ratios at the ER.To solve the formulated non-convex problem,we first find the optimal solution to the energy beamforming matrix,and then propose an iterative modified Newton method to solve the power splitting ratios optimization problem.Numerical results show that our proposed ER architecture and the joint optimization algorithm can exploit the nonlinearity of EH,and thus significantly improve the harvested DC power at the ER,as compared to the conventional designs.(2)We study the fairness beamforming design in the multiuser MISO WPT system under the non-linear EH model.Under this setup,we first find that the conventional optimal multi-beam energy transmission is not optimal any longer under the practical non-linear EH models.Then,we propose a novel time-division energy beamforming by jointly optimizing the beamforming matrices and the corresponding time duration to maximize the minimum harvested DC energy among all the ERs.To solve the formulated non-convex min-DC-energy maximization problem,we propose an efficient solution by using the techniques of alternating optimization and successive convex approximation(SCA).Numerical results show that our proposed design indeed improved the performance especially when the received RF power falls in the convex region for RF-to-DC conversion.(3)We investigate the channel estimation and analog beamforming design in multiuser MISO WPT system with single-tone signal transmission.On one hand,we propose a channel estimation method based on energy feedback by minimizing the mean square error(MSE)of the RF power and its measured value,in which the closed-form solution is obtained.On the other hand,due to the single-tone limitation,the optimal multi-beam transmission cannot be implemented in this system.Then,we proposed a time-division single-beam transmission,in which the energy beamforming is designed by adopting eigenvalue decomposition and Gaussian randomization together with time-sharing.And thus,the minimum harvested DC among all the ERs is maximized.Numerical results show that our proposed channel estimation method has a high accuracy,and the designed time-division single-beam transmission can almost achieve the same performance as compared with the optimal multi-beam transmission.(4)We consider an Intelligent Reflecting Surface(IRS)-assisted wireless powered over-the-air computation system,where there are two slots for wireless power transfer in the downlink and over-the-air computation in the uplink,respectively.Then,we formulate the computation rate maximization problem under the power constraints,transmit beamforming constraints,and passive beamforming constraints.We find that this non-covex optimization problem can be di vided into the min-DC-energy maximization in the downlink and min-SNR(signal-to-noise ratios)maximization in the uplink.Next,we propose the corresponding algorithms to solve these two problems,respectively.Numerical results show that our system design outperforms than the system without IRS or with random reflecting phases.In this thesis,by exploiting the nonlinearity of the EH model,our proposed ER architecture,beamforming design,and the channel estimation method can significantly improve the erengy conversion efficiency in WPT system,which possesses both the important research and practical value.And our designed IRS-assisted wireless powered over-the-air computation system as well as the discussion between the energy harversting and computation rate can provide the reference to the future work.