Cooperation and resource optimization in wireless networks
University of Delaware
Wireless technologies have been widely used to provide the communication functionality to various networking applications. However, the resources available to these applications are typically limited due to the fact that network devices are constrained by battery power, storage capacity and/or computational capabilities, and the fact that neighboring network devices share the common wireless spectrum with limited bandwidth. As a result, enabling cooperation among distributed network devices, and jointly optimizing the use of available resources become critical to wireless networks. In this dissertation, we study cooperation and resource optimization in various wireless networking scenarios including: ad hoc networks, underwater acoustic networks, wireless sensor networks and cellular networks. We aim to design efficient cooperation and resource optimization strategies to optimize system resource usage, and to maximize network performance. For ad hoc networks, we study the use of cooperative communications to combat various fading effects in wireless channels. We design the Cooperative Source Routing (CSR) protocol to translate the physical-layer gain of cooperative communications into improved network-level performance. By cooperatively transmitting both route-request and route-reply control protocol data units (PDUs), CSR is capable of exploring multi-hop Virtual Multiple-Input Single-Output (VMISO) paths with high cooperative diversity. For underwater acoustic networks (UWANs), we study the problem of transmission scheduling. We consider the issue of Spatio-Temporal Uncertainty attributed to the long propagation delay of acoustic signals, and specify constraints for conflict-free data transmissions and receptions in UWANs. We also design scheduling algorithms that exploit the long propagation delay of underwater acoustic signals, and facilitate concurrent transmissions and receptions to improve network throughput. For wireless sensor networks (WSNs), we study the problem of extending the lifetime of WSNs indefinitely . To address this issue, we make use of the wireless energy transfer (WET) technology, and propose the paradigm of Qi -ferries (QiFs). QiFs are equipped with wireless energy transmitters, and periodically rove a WSN to wirelessly charge the sensors' batteries. We investigate two tightly coupled issues of how QiFs plan the charging tours to ensure that no sensor exhausts its battery, and how the sensors adjust their power consumption (by employing different routing schemes) to better accommodate the QiFs' charging activities. For cellular networks, we consider the scenario of a group of neighboring mobile users (MUs) cooperating to download a video stream. In such a scenario, each MU requests a subset of the streaming contents over the MU's cellular link and shares the received contents with the neighboring MUs in a Peer-to-Peer (P2P) manner via a wireless local area network. We consider three issues here: the streaming data routing, the capacity of wireless local networks, and the MU's power consumption. Our objective is to determine an optimal P2P collaboration strategy that minimizes power consumption of the MUs. Lastly, for small-cell networks (SCNs), one of the prominent challenges is a lack of sufficient backhaul capacity to connect small-cell base stations (small-BSs) to a core network. To address this challenge, we exploit the effective application layer semantics of both spatial and temporal locality to reduce the backhaul traffic, by equipping small-BSs with a storage facility to cache contents requested by users. As the cache hit ratio increases, most of the MUs' requests can be fulfilled locally without incurring traffic over the backhaul. To make intelligent caching decisions, the mobility patterns of MUs must be carefully considered as MUs might frequently migrate from one small cell to another. We thus study the mobility-aware content caching problem, with the objective to maximize the caching utility. We show this problem is NP-complete, and develop a polynomial-time heuristic algorithm with bounded approximation ratio. The performance of the heuristic algorithm is evaluated using trace-based simulations.