Wireless sensor networks are evolving from specialized platforms to shared infrastructure for multiple applications. Shared sensor networks offer flexibility, adaptivity, and cost-effectiveness through dynamic resource allocation among different applications. Shared sensor networks face the critical need for optimizing Quality of Monitoring (QoM) subject to resource constraints. The emerging QoM optimization problems in shared sensor networks are computationally challenging due to their nonlinear, discrete, and dynamic nature. This project exploits a key property known as submodularity that many QoM attributes of physical phenomena exhibit. This project develops efficient and theoretically sound distributed approaches for QoM optimization, through a novel integration of submodular optimization in a market-based approach. It further studies online algorithms which can quickly adapt to network and application dynamics, partition-based algorithms that scale effectively for large-scale networks, and new optimization algorithms that can accommodate the optimization of energy consumption and heterogenous networks. Expected results of this project include theory, algorithms, and software for managing and optimizing a new generation of integrated networked sensing systems with high societal and environmental impact. The project also deploys an integrated sensor network for environmental monitoring in Tyson Research Center of Washington University for environmental and ecological research. This study promotes interdisciplinary collaboration with environmental and biological scientists, as well as outreach activities for high-school students.
Wireless sensor networks prevail in human life for various applications such as building monitoring, hazard monitoring, air/water quality monitoring and control, process monitoring and control. As many of these applications can share a single wireless sensor network, it is critical to optimally allocate the applications in a shared network to maximize the overall Quality of Monitoring that, in turn has direct impact on the public welfare. An important finding of this project is the development of a game-theoretic approach to distributed allocation of applications in a shared wireless sensor network that maximizes the overall Quality of Monitoring under the network resource constraints (e.g., in terms of memory and network bandwidth). We first transform the optimal application allocation problem to a submodular game and then develop a decentralized algorithm that only employs localized interactions among neighboring nodes. Previous solutions to this challenging application allocation problem are centralized in nature, limiting their scalability and robustness against network failures and dynamics. In contrast, our developed approach being distributed is highly scalable and robust against network dynamics. Simulations based on three real-world datasets demonstrate that our algorithm is competitive against a state-of-the-art centralized algorithm while scaling effectively with network size. We have also proposed a set of distributed near-optimal algorithms for allocating channels in wireless sensor networks. For wireless sensor networks in process monitoring and control applications, WirelessHART is a prominent standard. These applications impose stringent requirements on real-time and reliability guarantees in WirelessHART networks. The results and findings of this project include the development of an end-to-end delay analysis real-time flows for sensing and control in WirelessHART networks. The proposed delay analysis can be used to quickly assess the schedulability of real-time flows with stringent requirements on both high reliability and network latency under graph routing. We have implemented and evaluated our results on our existing wireless sensor network testbed at Washington University in St. Louis. Since a WirelessHART network is usually shared by many feedback control loops for monitoring and control purposes, it is critical to optimize the overall control performance in these networks. A critical finding of this project is a scheduling-control co-design approach for sampling rate selection to optimize the control performance. We studied four approaches to solve this challenging problem: a subgradient method, a simulated annealing based penalty method, a polynomial-time greedy heuristic method, and a gradient descent method based on a new delay bound that is convex and differentiable. We performed in-depth evaluation of these algorithms and gained valuable insights on their performance. In this project, we have also studied the lifetime maximization problem under high fault-tolerance requirement in industrial setting for WirelessHART networks. Power management is one of the challenging problems in wireless communication due to limited battery power of the networked devices. By carefully studying the transmission scheduling and energy consumption model in WirelessHART networks, we proposed near-optimal algorithms to construct maximum lifetime graph routes subject to fault tolerance requirements of WirelessHART networks.
