Performance Analysis And Optimization Of Wireless Powered Communication Systems

With the rapid development of the fifth generation mobile communication technology,various of wireless communication networks(WCN)such as internet of things(IoT),and wireless sensor networks has emerged in human society.The following explosive growth in the number of lowpower wireless communication devices,such as wireless sensor nodes,mobile phones,and medical implants,have raised great challenge in network designing: How to efficiently charge these devices wirelessly? To tackle this issue,wireless power transfer(WPT)technology based on radio frequency(RF)signals/optical signals has become promising solution to prolong the lifetime of WCNs.Since the RF/optical signals are fully controllable,the WPT technology is considered more reliable and efficient comparing to the conventional approach that harvests energy through ambient resources such as solar and wind.Therefore,the research of the combination of WPT technology with WCNs has gain considerably attention in recent years.As a result,network architectures such as wireless powered communications networks(WPCN),simultaneous wireless information and power transfer(SWIPT),and simultaneous lightwave information and power transfer(SLIPT)are proposed.In order to facilitate the application of the WPT technology into the WCNs.we present a comprehensive investigation on the system ergodic capacity,harvested energy,energy efficiency,block error rate(BLER),and the optimal beamforming designing problem,respectively.We first study the impact of Rician channels on the system ergodic capacity of a multi-input multi-output(MIMO)WPCN.Specifically,two separate cases are investigated according to the channel state information(CSI)at the energy transmitter,namely,the partial CSI case and no CSI case.In both cases,we derive the upper bound and lower bound for the system ergodic capacity.Besides,we investigate the capacity behavior in both high and low signal to noise(SNR)regimes,respectively.In addition,in order to maximize the capacity,we derive the optimal time allocation factor that decides the duration of energy harvesting process and information transmission process,with respect to each SNR regime.The findings suggest that a stronger line of sight(LOS)component in Rician channel benefits the system capacity.And,deploying more antennas at either node in the system will greatly increase the system performance.Next,we investigate the energy efficiency of the massive MIMO WPT system that adopts a mixed analog to digital conversion(ADC)antenna architecture base station(BS).Specifically,based on the mixed-ADC antenna architecture at BS,we derive the minimum mean square estimation(MMSE)of the channel matrix.Then,with conjugate beamforming,we derive the total harvested energy of all users.Additionally,we investigate the optimal ratio allocation for the high/low resolution ADCs by solving the energy efficiency optimization problem.The results show that the optimal ADC ratio allocation strategy is binary,which means that either fully high resolution ADCs or fully low resolution ADCs are required so as to obtain the maximum energy efficiency.Besides,our findings suggest that the decision threshold of the binary solution depends heavily on the adopted energy consumption model.Then,the average block error rate(BLER)of a cooperative SWIPT system with finite block length transmission was investigated.Specifically,we derive the integral expression of the average BLER,based on which the numerical analysis of the system performance is studied.In addition,the tight lower bound on the system BLER is derived.Moreover,we derive the approximated BLER under high-SNR assumption.The results show that the system BLER can be reduced with smaller packet length.Besides,in order to obtain better BLER performance,the relay positioning is discussed.Finally,we study the optimal beamforming design for an in-door two-user MIMO SLIPT system.Through solving the max-min user SNR optimization problem under specific energy harvesting constraints and light emitting diode(LED)input power constraints,we obtain the optimal beamforming design for the system.Besides,two base line beamforming approaches based on zero forcing(ZF)precoding are presented as comparisons.Our findings suggest that the proposed optimal beamforming strategy outperforms the ZF base line beamforming strategies,which have been widely adopted in the research field of visible light communication(VLC)systems with energy harvesting capability.