Efficient and robust protocols for distributed control and optimization of ubiquitous wireless networks

多跳无线网络，特别是那些需要低成本基础设施的网络，为各种关键民用和军用应用中的无线宽带接入提供了经济的解决方案。尽管在过去的几年中，这些无线网络的协议工程相关的积极的研究活动，其成功的实际实施是相当有限的。因此，更基础的研究，可以指导这些网络的低复杂性和实际可实现的网络协议的设计是非常必要的。为此，本研究的主要目标是建立一个统一的理论，系统建模，分析和协议工程，导致实际的网络协议的多跳无线网络的发展。具体来说，我们建议考虑设计的通信方案和网络协议的协议栈，使我们能够成功地实现这些物理无线系统的所有层。我们的目标是了解最大可实现的吞吐量区域方面的性能限制，考虑现实的物理层模型和先进的通信技术，超越了简单的成对干扰模型，认为大多数现有的作品在文献中。我们将开发跨层控制协议，在不同的时间尺度上适当地工作。这些多层协议应该能够利用层之间的交互来增强网络性能，同时最大限度地减少分层模块化的损失，同时它们可以很容易地以分布式方式实现。此外，它们应该在吞吐量、延迟和数据包丢失等几个关键性能指标方面保持高效，并且对不同形式的不确定性（例如，在网络拓扑和信道状态信息中）。 最后，拟议的研究将考虑异构无线电接入技术以及动态频谱接入能力的无线设备，这是未来的自适应无线系统的重要方面。 这项研究将有助于生产具有成本效益的无线技术，并提供先进无线技术领域的HQP，这有助于促进我国的技术转让和经济发展。

Multihop wireless networks, especially those requiring low-cost infrastructure, provide an economical solution for wireless broadband access in various critical civilian and military applications. In spite of active research activities related to protocol engineering for these wireless networks in the past several years, their successful practical implementations are rather limited. Therefore, more fundamental research studies that can guide the design of low-complexity and practically-implementable network protocols for these networks are strongly needed. To this end, the main goal of this proposed research is to build a unified theory for system modeling, analysis and protocol engineering that leads to developing of practical network protocols for multihop wireless networks. Specifically, we propose to consider the design of communication schemes and network protocols in all layers of the protocol stack that enables us to successfully realize these physical wireless systems. We aim to understand the performance limits in terms of maximum achievable throughput region considering realistic physical layer models and advanced communication techniques beyond simplistic pair-wise interference models considered by most existing works in the literature. We will develop cross-layer control protocols that work appropriately over different time scales. These multi-layer protocols should be able to exploit the interactions among layers to enhance network performance with minimal loss of layering modularity while they can be easily implemented in a distributed fashion. In addition, they should be efficient with respect to several key performance metrics such as throughput, delay and packet loss and robust to different forms of uncertainties (e.g., in network topology and channel state information). Finally, the proposed research will consider heterogeneous radio access technologies as well as dynamic spectrum access capability of wireless devices, which are important aspects of a future adaptive wireless system. This proposed research will contribute to produce cost-effective wireless technologies and provide HQP in the advanced wireless technology sector, which helps boost the technology transfer and economic development for our country.