Abstract
Wireless powered communication networks (WPCNs) are commonly analyzed by using the linear energy harvesting (EH) model. However, since practical EH circuits are non-linear, the use of the linear EH model gives rise to distortions and mismatches. To overcome these issues, we propose a more realistic, nonlinear EH model. The model is based upon the error function and has three parameters. Their values are determined to best fit with measured data. We also develop the asymptotic version of this model. For comparative evaluations, we consider the linear and rational EH models. With these four EH models, we investigate the performance of a WPCN. It contains a multiple-antenna power station (PS), a signal-antenna wireless device (WD), and a multiple-antenna information receiving station (IRS). The WD harvests the energy broadcast by the PS in the PS-WD link, and then it uses the energy in the WD-IRS link to transfer information. We analyze the average throughput of delay-limited and delay-tolerant transmission modes as well as the average bit error rate (BER) of binary phase-shift keying (BPSK) and binary differential phase-shift keying (BDPSK) over the four EH modes. As well, we derive the asymptotic expressions for the large PS antenna case and the effects of transmit power control. Furthermore, for the case of multiple WDs, we optimize energy beamforming and time allocation to maximize the minimum rate of the WDs. Finally, the performances of four EH models are validated by Monte-Carlo simulations.
Highlights
2) To comparatively evaluate these four models, we investigate the throughput of the Wireless powered communication networks (WPCNs) (Fig. 3) and bit error rates (BER) of binary phase-shift keying (BPSK), binary differential phase-shift keying (BDPSK) modulations
We show that the optimal energy beamformer significantly outperforms omni-directional beamforming
We show that NLEH, AM, and RM models accurately reach the saturation state of practical energy harvesting (EH) circuits, but the LM
Summary
I NTERNET of Things (IoT) is a network to connect people, processes, data, and things. The sigmoid model posits a logistic transfer function between the input and output powers [31] It captures the saturation characteristic of practical circuits but assumes zero sensitivity. In [42], with the sigmoid nonlinear EH model, the time allocation and power control of a multi-user WPCN system are jointly optimized to maximize the minimum individual throughput. In this scheme, the uplink WDT is supported through time division multiple access (TDMA) with multi-input multi-output (MIMO) transceivers. NEW ENERGY HARVESTING MODEL As mentioned in Section I-B, there are two limitations in the practical EH circuits: First, for large input powers, the output power exhibits a saturation plateau.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
