NeTS: Small: Collaborative Research: Compressed Network Tomography and Data Collection in Large-Scale Wireless Sensor Networking

我们的物质世界呈现了一套令人难以置信的丰富的观察模式。无线传感器网络(WSN)的最新进展使人们能够以前所未有的高空间密度和长时间连续监测各种物理现象，从而为众多科学努力打开了令人兴奋的新机会。由于传感器节点是无人值守和电池供电的，因此在大规模环境无线传感器网络的部署中，间接测量的网络监测/层析成像(S)和节能是至关重要的。因此，一个可行的节能网络监测和数据采集框架对于显著改善无线传感器网络的管理/运营和降低其部署成本具有重要意义。本项目基于压缩感知技术(CS)的最新突破，通过理论和实证相结合的方法，研究大规模室外无线传感器网络的节能网络监测/层析成像和数据采集。本课题研究无线传感器网络拓扑层析技术在无线链路动态衰落和干扰下的动态路由问题。本项目的目标是为运行在高噪声通信环境中的真实世界的无线传感器网络开发一种新颖而严格的拓扑层析成像框架。针对汇点接收到的完全间接测量和不完全间接测量，设计了动态路由拓扑恢复算法(S)。从分析和实验两个方面研究了层析成像方法的准确性。所开发的WSN拓扑层析框架不仅对于实际中的WSN的路由改进、拓扑控制、热点消除和异常检测是必不可少的，而且对于新兴的基于CS的数据收集也是必不可少的。该方法扩展了现有的CS技术，形成了大规模无线传感器网络中网络层析成像和数据采集的统一框架，并在此基础上开发了能量高效的无线传感器网络拓扑层析成像和数据采集协议族。开发的框架和协议集将在一个位于丘陵分水岭的真实环境中的无线传感器网络试验台上进行验证和评估。该项目旨在为大规模无线传感器网络创造一种新的优化设计、开发和管理/运营的范例，以显著延长其寿命。这将在不久的将来大幅降低用于科学、民用、国家安全和军事目的的大规模无线传感器网络部署的高昂成本。该项目为本科生和研究生创建了一种跨学科的教育实践，通过真实世界的无线传感器网络试验床的实践经验。外展活动包括夏令营和为使用WSN试验床的在校学生开展的科学项目。

Our physical world presents an incredibly rich set of observation modalities. Recent advances in wireless sensor networks (WSNs) enable the continuous monitoring of various physical phenomena at unprecedented high spatial densities and long time durations, hence opening exciting new opportunities for numerous scientific endeavors. Since sensor nodes are unattended and batterypowered, network monitoring/tomography from indirect measurements at the sink(s) and energy conservation are critical in the deployment of large-scale environmental WSNs. Therefore, a viable framework for energy-efficient network monitoring and data collection is fundamentally important to significantly improve WSN management/operations and reduce its deployment costs.This project investigates the energy-efficient network monitoring/tomography and data collections in large-scale outdoor WSNs, based on the recent breakthrough of compressed sensing (CS) through an integrated theoretical and empirical approach. The project studies WSN topology tomography for dynamic routing under wireless link dynamics due to channel fading and interference. The objectives of this project are to develop a novel and rigorous framework of topology tomography for real-world WSNs operated in highly noisy communication environments. Dynamic routing topology recovery algorithms are devised for both complete indirect measurements and incomplete indirect measurements received at the sink(s). The accuracy of the tomography approach is studied both analytically and empirically. The developed WSN topology tomography framework can be essential not only for WSN's routing improvement, topology control, hot spot elimination, and anomaly detection in practice, but also for emerging CS-based data collection. This approach extends the current CS technology to form a unified framework for network tomography and data collection in large-scale WSNs, upon which energy-efficient WSN topology tomography and data gathering protocol suite is developed. The developed framework and protocol suite will be validated and evaluated in a real-world environmental WSN testbed in a hilly watershed.The project intends to create a new paradigm of optimal design, development, and management/operations for large-scale WSNs to significantly extend their lifetime. This would lead to a substantial reduction of the prohibitive cost of large-scale WSN deployments for scientific, civic, national security, and military purposes in the near future. The project creates an interdisciplinary educational practice for both undergraduate and graduate students through hands-on experience with a real-world WSN testbed. The outreach includes summer camps and scientific projects for school students using the WSN testbed.