|dc.description.abstract||The limited life time of batteries is a crucial issue in energy-constrained wireless communications. Recently, the radio frequency (RF) wireless energy transfer (WET) technique has been developed as a new practical method to extend the life time of wireless communication networks. Inspired by this, wireless-powered communication network (WPCN) has attracted much attention. Therefore, in this thesis, we consider practical WET and wireless-powered information transmission in WPCNs.
First we investigate a WPCN with two nodes, in which an access point (AP) exchanges information with a wireless-powered user. The user is assumed to have no embedded energy supply and needs to harvest energy from RF signals broadcast by the AP. Differing from existing work that focuses on the design of wireless-powered communication with one-way information flow, we deal with a more general scenario where both the AP and the user have information to transmit. Considering that the AP and user can work in either half-duplex or full-duplex mode as well as having two practical receiver architectures at the user side, we propose five elementary communication protocols for the considered system. Moreover, we define the concept of a throughput region to characterize the tradeoff between the uplink and downlink throughput in all proposed protocols. Numerical simulations are finally performed to compare the throughput regions of the proposed five elementary protocols.
To further the study on WPCN, we investigate a wireless-powered two-way relay system, in which two wireless-powered sources exchange information through a multi-antenna relay. Both sources are assumed to have no embedded energy supply and thus first need to harvest energy from the radio frequency signals broadcast by the relay before exchanging their information via the relay. We aim to maximize the sum throughput of both sources by jointly optimizing the time switching duration, the energy beamforming vector and the precoding matrix at the relay. The formulated problem is non-convex and hard to solve in its original form. Motivated by this, we simplify the problem by reducing the number of variables and by decomposing the precoding matrix into a transmit vector and a receive vector. We then propose a bisection search, a 1-D search and an iterative algorithm to optimize each variable. Numerical results show that our proposed scheme can achieve higher throughput than the conventional scheme without optimization on the beamforming vector and precoding matrix at the relay.
Due to the high attenuation of RF energy over a long distance, RF based wireless-powered communication is usually designed for low-power scenarios, e.g., wireless-powered sensor networks. Recently, magnetic induction (MI) based WET has been proposed to wirelessly transfer a large amount of energy. Inspired by this, we investigate MI based WET in WPCN. Specifically, we study a MI based wireless-powered relaying network, in which a MI source transmits information to a MI destination, with the help of a MI based wireless powered relay. We propose four active relaying schemes, which consider different relaying modes and different energy harvesting receiver architectures at the relay. We then aim to maximize the end-to-end throughput of each scheme by using a bisection search, a water-filling algorithm, a Lagrange multiplier, quasi-convex programming and an iterative algorithm. We compare the proposed active relaying schemes with passive relaying. Numerical results show that the proposed relaying schemes with a decode-and-forward relaying mode significantly improve the throughput over passive relaying.||en_AU|
|dc.publisher||University of Sydney||en_AU|
|dc.publisher||Faculty of Engineering & Information Technologies||en_AU|
|dc.publisher||School of Electrical and Information Engineering||en_AU|
|dc.publisher||Centre of Excellence in Telecommunications||en_AU|
|dc.title||Wireless Powered Communication Networks||en_AU|
|dc.type.pubtype||Master of Philosophy M.Phil||en_AU|
|Appears in Collections:||Sydney Digital Theses (Open Access)|