NeTS: Small: Predictable Optimization of Opportunistic Communication

点击翻译按钮获取中文摘要

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5). Opportunistic communication leverages communication opportunities arising by chance to provide significant performance benefit or even enable communication where it would be impossible otherwise. This project develops algorithms, techniques, and protocols that optimize opportunistic communication to achieve good, predictable performance in wireless mesh networks and vehicular networks.A key challenge involved is how to systematically optimize opportunistic communication to achieve good predictable wireless performance. The research addresses the challenge by making three major contributions. First, the PIs develop novel robust optimization techniques that systematically optimize opportunistic communication even in the presence of high uncertainty. Second, they design efficient models, measurement and inference techniques, and prediction algorithms to obtain the required inputs to the optimization algorithms. Third, they exploit inter-flow coding opportunities arising from multi-flow diversity to further enhance the efficiency of opportunistic communication. To demonstrate the effectiveness of the approaches, the PIs design, implement, and experiment in a wireless mesh network and a vehicular network testbed deployed at UT Austin.The project produces publications in leading network conferences and journals, and software that is publicly available online. The resulting techniques and software significantly advance wireless mesh network and vehicular network technology, and benefit other wireless network environments. The research results are also integrated into undergraduate and graduate curricula as well as outreach activities.