Wireless networks, by virtue of their untethered nature, ease of set-up, and mobility support, have been heartily welcomed into our daily work and lives. Crucial to the continuing success and even wider deployment of wireless networks is automatic management of these networks.
The goal of this research is to investigate two fundamental questions of fault management in wireless networks: (1) What are suitable management architectures and what data/measurements need to be collected for effective management? (2) What are suitable techniques for fault management? To be focused, this research mainly investigates two ?extreme? forms of wireless networks -- wireless LANs (WLANs) and sensor networks. More specifically, it proposes a novel management architecture and develops a series of fault management techniques for WLANs. It also develops a suite of diagnostic tools for static and mobile sensor networks, respectively. The techniques for the various management tasks utilize statistical analysis, optimization, stochastic modeling, and machine learning techniques, and will be evaluated using a combination of analysis, simulation and experiments. Although necessarily focused on a subspace in the large realm of wireless network management, by synthesizing lessons learned from this subspace, this project aims to provide insights into wireless network management in general.
The outcome of this project will greatly advance the state of the art in wireless network management, potentially impacting many aspects of our daily lives. The education plan includes developing one undergraduate lab course on WLANs, and one graduate-level course on wireless network measurement and inference.
In this project, we aim to automate wireless network fault management to assure greater stability and longer service life, which is crucial to the continuing success and even wider deployment of wireless networks. Our efforts are mainly on two aspects: management architectures and techniques. We primarily focus on WiFi and wireless sensor networks since they represent two 'extremes': WiFis are single-hop networks with infrastructure support; while sensor networks are multi-hop networks, with no infrastructure support and more stringent resources constraints. By synthesizing and extending lessons learned from these two 'extreme' networks, we expect to shed light on managing wireless networks in general. Over the duration of the project, we have developed a suite of novel fault management architectures and techniques for both WiFi and wireless sensor networks. Many of the techniques can be applied broadly to other types of wireless networks. Since wireless networks have become more and more widely deployed and our dependence on them grows, we expect our findings will benefit many aspects of our daily life and work. The grant has partially supported six Ph.D. students. Three of the students (one female) have graduated with doctorial degrees. The REU supplement to this grant has supported eight REU students (one Hispanic) over the years, leading to two publications. The PI has developed a graduate-level course on wireless networks and a graduate/undergraduate lab course on wireless sensor networks. These two courses have been offered multiple times, training tens of graduate and undergraduate students. In addition, the PI has presented the research to over 100 visiting high-school students. A website dedicated to this project is at http://nlab.engr.uconn.edu/currentprojects_awnm.html.
