NeTS-WN: Collaborative Research: Channel-Aware Distributed Scheduling for Optimal Throughput and Latency: A Unified PHY/MAC Approach

无线自组织网络的设计在无线通信中面临着许多独特的挑战，包括1)邻居中活动链路之间的共信道干扰，以及2)衰落信道上的时变信道条件。实验数据表明，在许多现实场景中，衰落效应往往会对MAC层产生不利影响，衰落时间尺度和MAC层之间的耦合要求统一的PHY/MAC设计。由于自组织通信的分布式本质，几乎没有工作来开发用于吞吐量最大化的信道感知的分布式调度。目前还没有针对延迟约束下的实时业务的信道感知调度的系统研究。本项目的一个主要目标是填补这一空白，并为无线自组织网络中针对弹性业务和非弹性业务的信道感知的分布式调度奠定理论基础。以开发一个统一的PHY/MAC优化框架为目标，提出的研究包括三个推力。前两个推力研究弹性流量的分布式机会调度，分别从以网络为中心和以用户为中心的角度关注吞吐量最大化。第三个重点是针对实时业务的显式时延约束下的网络模型的信道感知调度。这项研究综合运用了调度、随机优化、博弈论和控制论中的基本工具。这个项目将为探索无线自组织通信的信道感知分布式调度开辟一条新的途径。PIS预计，所提出的工作将最终形成新的无线自组织网络的基本理论和先进的设计方法，并将对包括无线局域网和无线Mesh网络在内的许多无线应用产生重大影响。除了技术影响，拟议研究的更广泛影响还包括教育因素。

The design of wireless ad-hoc networks faces a number of unique challenges in wireless communications including 1) co-channel interference among active links in a neighborhood, and 2) time-varying channel conditions over fading channels. Experimental data reveals that, in many realistic scenarios, fading effects can often adversely affect the MAC layer, and the coupling between the timescales of fading and MAC calls for a unified PHY/MAC design. Due to the distributed nature of ad hoc communications, little work has been done to develop channel-aware, distributed scheduling for throughput maximization. There are virtually no systematic studies on channel-aware scheduling for real-time traffic under latency constraints. A principal goal of this project is to fill this void and build a theoretic foundation for channel-aware, distributed scheduling in wireless ad-hoc networks, for both elastic traffic and inelastic traffic. With the goal of developing a framework for unified PHY/MAC optimization, the proposed research consists of three thrusts. The first two thrusts investigate distributed opportunistic scheduling for elastic traffic and focus on throughput maximization from network-centric and user-centric perspectives, respectively. The third thrust focuses on channel-aware scheduling for network models under explicit delay constraints, for real-time traffic. The proposed research draws on a combination of fundamental tools in scheduling, stochastic optimization, game theory, and control theory. This project will open a new avenue for exploring channel-aware distributed scheduling for ad-hoc communications.The PIs expect that the proposed work will culminate in the formulation of both new fundamental theories and advanced design methodologies for wireless ad hoc networks, and will have a significant impact on many wireless applications including wireless LANs and wireless mesh networks. In addition to the technical impacts, the broader impacts of the proposed research also include educational elements.