Electronic Theses and Dissertations for Graduate School
Add My Work
Author Last Name
Energy and Signal Cooperation in Competitive Wireless Networks with Energy Harvesting
Restricted (Penn State Only)
Doctor of Philosophy
Date of Defense:
August 29, 2017
Aylin Yener, Dissertation Advisor
Aylin Yener, Committee Chair
Viveck Cadambe, Committee Member
Chris Giebink, Committee Member
Soundar Kumara, Outside Member
Energy harvesting is a green communications paradigm where the nodes harvest their energy from external and intermittent sources such as solar radiation and other wireless nodes' transmission. This green approach promotes an efficient utilization of the available resources, but also calls for careful management of these resources. Cooperation between energy harvesting nodes further improves network-wide performance and ensures a more balanced distribution of the network resources, notably, energy. This includes signal cooperation and energy cooperation. Signal cooperation in a wireless network entails nodes assisting other nodes by relaying their data. Thus, it requires the relaying nodes to take into account the data they receive and allocate their available energy accordingly. This introduces a data aspect to the optimization of energy harvesting networks which is often not addressed in studies that do not focus on cooperation. Energy cooperation improves network performance by letting nodes share their available energy with each other. This yields a fair distribution of energy in the network and counteracts energy starvation in the network. While the majority of works on energy and signal cooperation are built upon the assumption that all nodes will work towards a common goal in an altruistic fashion, this assumption may not be realistic in real world communication scenarios. In this dissertation, we study cooperation in energy harvesting networks and propose incentivization schemes where nodes can trade energy cooperation for signal cooperation and vice versa. We start with signal cooperation in energy harvesting networks. The network models that we consider range from a simple network of an energy harvesting transmitter and a receiver, to an energy harvesting multi-way relay channel. We consider finite batteries and finite data buffers and formulate the (sum) throughput maximization problem in offline and online settings. We consider quality of service requirements such as delay constraints, and fading as well. We identify numerical and algorithmic optimal solutions to the (sum) throughput maximization problem. We show that we can decouple the problem into energy allocation and data allocation subproblems. We extend directional waterfilling to solve the former and propose an optimal solution for the latter that is based on induction. We extend this approach to larger networks including the two-way and two-way relay channels. In addition, we address the cases of continuous energy arrivals as well as energy harvesting cognitive radios. We next bring energy cooperation into the picture and study a competitive scenario with selfish nodes that aim to maximize their own utilities only. We envision the nodes to offer signal cooperation in return for energy cooperation and vice versa, and take into account the impact of their actions on other nodes' utilities. We consider a two-hop network with energy transfer from the sources to the relays in exchange for the relaying of the sources' data. We study several game theoretical models such as noncooperative and Stackelberg games, and Vickrey auctions. We next study a network of transmitters and receivers where the transmitters can offer energy to the receivers. We consider one-to-one and one-to-many matching games where each transmitter can be matched with one or multiple receivers. For both cases, we identify stable matchings that maximize the sum rate and show the influence energy transfer offers on the matchings.
Login using your Penn State access account to view the paper.