Resource Management for High Performance Reliable Wireless Networking

本课题研究的核心是开发无线网络的资源管理机制，以满足用户需求，提供可靠的服务。网络必须适应广泛的应用。这些应用程序具有不同的资源要求和不同的性能预期。管理无线网络中的资源是具有挑战性的，因为它们有带宽、能量和处理限制，而且更容易出错。用户期望有可靠、及时的服务交付。但是，这些服务可能发生在不同的有线和无线网络上。因此，在无线网络上提供高质量和高可靠性的网络服务是一个具有挑战性的问题。我的研究涉及为无线网络设计资源管理协议。本文采用模拟、分析和实证研究相结合的方法对这些机制的绩效进行评价。在这一应用中提出了以下研究方向。在无线网络中，共享网络接入的竞争是满足用户需求和提供可靠服务的瓶颈。1)在第一个项目中，将为多跳多分组接收自组织无线网络设计和评估新的流量控制和调度机制。关于物理层的多分组接收和多分组传输的有前景的新发展具有改善链路层的干扰相关损耗的潜力。然而，在以前的工作中，我们发现了将现有链路层接入协议与此技术结合用于多跳无线网络时存在的瓶颈问题。通过我们提出的工作，我们将能够充分发挥节点扩展接收能力的潜力，并极大地减少共享环境中链路层的瓶颈问题。2)链路层接入的另一个关键问题是城市无线网络中的噪声干扰问题，其中许多无线网络(以及具有干扰信号的其他设备)共存。在第二个项目中，解决了节点如何对这种外部干扰进行分类并自适应地与邻居协调这些知识以改善通信的问题。在未来，更多的网络和技术将共存。因此，需要项目2中设计的机制来允许节点适应不同的环境，提高可靠性性能。3)在以前的工作中，我们开发了一个控制框架，用于在基于竞争的链路层无线协议中提供公平的资源访问。我们还为可靠的数据传输设计了虚拟拓扑。在项目3中，这项工作将结合起来，以产生全面的数据交付机制，包括路由、可靠传输和受控链路访问。机动性和公平性也将得到考虑。用于放置资源的虚拟拓扑在移动云计算等应用中发挥着重要作用，在移动云计算中，及时访问和放置信息至关重要。为了向应用程序提供无缝服务，需要通过不同的有线和无线网络可靠地传输数据。这些都是项目3中拟议工作的贡献。总而言之，联网的最终目标是提供可靠的服务，以满足使用网络的许多应用程序的需求。在无线网络中，处理对传输介质的访问的链路层是良好性能的瓶颈。这项研究解决了这一层的问题，提供了显著提高性能的机制，包括提高可靠性、增加吞吐量、减少延迟和增加公平性。

The core of my research is to develop resource management mechanisms for wireless networks in order to meet user requirements and provide reliable service. Networks must accommodate a wide range of applications. These applications have different resource requirements and different performance expectations. Managing resources in wireless networks is challenging because they have bandwidth, energy and processing constraints and they are more error-prone. Users expect to have reliable, timely delivery of services. However, these services may occur over heterogeneous wired and wireless networks. Therefore, providing network services over wireless networks with quality and reliability is a challenging problem. My research involves designing protocols for resource management for wireless networks. I use simulation, analysis and empirical studies to evaluate the performance of the mechanisms. The following research directions are proposed in this application. In wireless networks, the competition for shared access to the network is a bottleneck to meeting user requirements and providing reliable service. 1) In the 1st proposed project, new flow control and scheduling mechanisms will be designed and evaluated for multi-hop multi-packet reception ad hoc wireless networks. Promising new developments regarding multi-packet reception and multi-packet transmission at the physical layer have the potential to improve interference-related loss at the link layer. However, in previous work, we identified bottleneck issues when existing link layer access protocols are used with this technology for multi-hop wireless networks. With our proposed work, we will be able to reach the full potential of the expanded reception capability of nodes and vastly reduce the bottleneck issues of the link layer in shared environments. 2) Another critical issue with link layer access is the problem of noise interference in urban wireless networks where many wireless networks (and other equipment with interfering signals) co-exist. In the 2nd project, the problem of how nodes classify this outside interference and coordinate this knowledge adaptively with their neighbours to improve communication is addressed. In the future, even more networks and technology will co-exist. Therefore, the mechanisms designed in Project 2 are needed to allow nodes to adapt to different environments and improve reliability performance. 3) In previous work, we have developed a control framework for providing fair access to resources in contention-based link layer wireless protocols. We have also designed virtual topologies for reliable data delivery. In Project 3, this work will be combined to produce comprehensive data delivery mechanisms that include routing, reliable transmission and controlled link access. Mobility and fairness will also be considered. Virtual topologies for placement of resources serve an important role in applications such as mobile cloud computing where timely access and the placement of information is of critical importance. Reliable delivery of data over different wired and wireless networks is required for seamless service to applications. These are the contributions of the proposed work in Project 3. In summary, the ultimate goal of networking is to provide reliable service that meets the needs of the many applications that use the network. In wireless networks, the link layer which handles access to the transmission medium, is a bottleneck to good performance. This research address the issues at this layer by providing mechanisms to significantly improve performance, including increased reliability, increased throughput, reduced delay and increased fairness.