报告题目:Characterizing Energy-Delay Tradeoff in Hyper-Cellular Networks with Base Station Sleeping Control
报告人: 牛志升教授（清华大学电子工程系清华信息科学与技术国家实验室）
报告时间: 5月30日（星期五）上午 11:00-12:00
地点: 信息学院四楼报告厅
摘要：
One of the key approaches to make the mobile communication networks more GREEN (Globally Resource-optimized and Energy-Efficient Networks) is to have the cellular architecture and radio resource allocation more adaptive to the environment and traffic variations, including making some lightly-loaded base stations (BSs) go to sleep. This is the concept of so-called TANGO (Traffic-Aware Network planning and Green Operation) and CHORUS (Collaborative and Harmonized Open Radio Ubiquitous Systems) published by the author earlier. To realize this, a new cellular framework, named hyper-cellular networks (HCN), has been proposed, in which the coverage of control signals is decoupled from the coverage of data signals so that the data coverage can be more elastic in accordance with the dynamics of traffic characteristics and QoS requirements. Due to this elasticity of HCN, some delay-insensitive users may have to experience some delay or other kind of QoS degradation when traffic load is high in order to save energy, i.e., energy can be traded off by some delay. The fundamental question then arises: how much energy can be traded off by a tolerable delay?
In this talk, we characterize the tradeoffs between energy consumption and service delay in a base station with sleep mode operations by queueing models. The base station is modeled as an M/G/1 vacation queue with setup and close-down times, where the base station enters sleep mode if no customers arrive during the close-down time after the queue becomes empty and it starts to setup when it sees N arriving customers during its sleep period. Several closed-form formulas are derived to demonstrate the tradeoffs between the energy consumption and the mean delay by changing the close-down time, setup time, and N. It is shown that the relationship between the energy consumption and the mean delay is linear in terms of mean close-down time, but non-linear in terms of N. The explicit relationship between total power consumption and average delay with varying service rate is also analyzed theoretically, indicating that sacrificing delay cannot always be traded for energy saving. In other words, larger N may lead to lower energy consumption, but there exists an N* that minimizes the mean delay. We also investigate the maximum delay for certain percentage of service, which is closely related to the mean delay. In summary, the closed-form tradeoffs cast light on designing BS sleep control policies which aim to save energy while maintaining acceptable quality of service.