No abstract

D2D communications underlaying cellular networks has become very popular recently. Interference between cellular users and D2D users is one of the tricky issues we need to solve. Another challenging issue is limited battery lives. In this thesis, we address these two issues together by introducing an energy harvesting (EH) assisted D2D model where the whole transmission process is divided into multiple time slots. At the beginning of every time slot, each D2D user will update the remaining energy level to the base station (BS) and the BS will then decide whether the D2D user should harvest energy or transmit data. The objective is to maximize sum throughput for all D2D users. To solve the problem, we first adopt the NOMAD algorithm, then propose a heuristic algorithm as sub-optimal solution. Numerical results show that our proposed algorithm can achieve almost the same sum throughput at a significantly smaller time cost.iii AcknowledgmentFirst of all, I would like to thank my supervisors, Alagan Anpalagan and Ling Guan, for their help and support. Every meeting has contributed to the progress of my research and thesis. Their patience and understanding is also what makes this journey more enlightening and rewarding. I also want to thank a fellow researcher in the lab, Dr. Waleed, for his helpful suggestions and all the insightful discussions we had. You will make an amazing professor one day.Second, I would like to extend my gratitude to the Department of Electrical and Computer Engineering and the School of Graduate Studies of Ryerson University for their financial supports and all the amazing opportunities they have provided for me.Finally, I would like to thank my parents and friends, without whose help and encouragement I would not be able to achieve my goal and finish my degree. I promise I will continue chasing my dreams and make you proud.

Device to device (D2D) communication is capable to address the increasing demand for data rates in fifth generation (5G) and beyond networks. However, D2D communication is usually convoluted with interference scenarios since both D2D users and cellular users share the same spectrum resources. Furthermore, D2D systems can trace back to limited battery life. The battery life problem is becoming more challenging with the exponential increase of devices in the future networks. Therefore, efficient resource allocation schemes need investigation to offer better quality of service for both cellular and D2D users under the constraints of interference and energy. In this paper, we address these two problems (interference and energy) simultaneously by efficiently allocating resources in energy harvesting‐assisted underlay D2D communication. We propose a deterministic model in which D2D users harvest energy only when required. We propose a resource allocation scheme, which jointly allocate resources and transmit power. We formulate an optimization problem with an objective to maximize sum throughput of D2D system while satisfying constraints on quality of service, power, and interference. To solve the problem, we adopt the nonlinear optimization by mesh adaptive direct search algorithm to obtain the suboptimal solution. We show the effectiveness of the proposed scheme in comparison with existing algorithms through simulation results.