Wireless Metropolitan Area Networks (Wireless MAN) are emerging as a highly promising Broadband Wireless Access (BWA) technology that has attracted much attention. Due to the increasing demand in wireless services associated with the scarcity of the battery power, how to design and implement an energy-efficient system with QoS support has become a dominating issue in Wireless MAN technology....[ Read more ]

Wireless Metropolitan Area Networks (Wireless MAN) are emerging as a highly promising Broadband Wireless Access (BWA) technology that has attracted much attention. Due to the increasing demand in wireless services associated with the scarcity of the battery power, how to design and implement an energy-efficient system with QoS support has become a dominating issue in Wireless MAN technology.

To fulfill these requirements, IEEE 802.16e-2005 (Mobile Broadband Wireless Access), the mobile version of WiMax, supports the Power Saving Mechanism (PSM) on the Medium Access Control (MAC) layer and provides multiple sleep modes to extend the battery life of portable devices. We provide simple but accurate analytical models, capable of calculating the sleep ratio, the system-level energy consumption, as well as the mean packet delay for IEEE 802.16e PSM evaluation. In addition, a comparative study has been conducted among these performance metrics and we investigate the energy-performance trade-off between different sleep mode types. Finally, we also investigate the impact of the system parameter settings to the performance and propose a heuristic algorithm on sleep mode selection as well as corresponding parameter tuning.

Since IEEE 802.16e PSM offers flexible operational modes, to demonstrate the potential benefits, we aim at designing a proper sleep mode operation scheme that has been left unspecified in the standard. The energy-performance trade-offs revealed by the previous performance evaluation also inspire us to conduct optimization issue towards IEEE 802.16e PSM. Therefore, we build up a theoretical framework based on the Markov Decision Processes (MDP) in this thesis. The MDP approach allows us to formulate stochastic constrained optimization problems that are capable of finding the best strategy for energy-efficient operation. Additionally, our methodology is investigated in several design scenarios, along with the evaluation on optimal policy distribution under different traffic loads. For the fulfillment of our study, we also put forward an implementation model that can efficiently conduct optimal sleep mode operation on portable devices in the runtime environment.

Finally, we outline several relevant open issues based on our research work. We believe our performance evaluation and MDP framework will contribute to both theoretical research and practical implementation in the energy-efficient design on mobile wireless broadband access systems.