Lecture Series on Information Processing in Sensor Networks

Wireless sensor networks have recently come into prominence because they hold the potential to revolutionize many segments of our economy and life, from environmental monitoring and conservation, to manufacturing and business asset management, to automation in the transportation and health-care industries, to battlefield awareness and other defense applications. Embedding sensors in the physics world and creating a network poses many new challenges, both because of bandwidth and energy limitations (untethered nodes are common), but also because tracking physical phenomena generates traffic patterns that are fundamentally different from the more familiar peer-to-peer networks. Those constraints dictate that sensor network problems are best approached in a holistic mannar, by jointly considering the physical, networking, and application layers and making major design trade-offs across the layers.

This set of lectures will examine how to design and analyze the implementation of information processing tasks on sensor netowrks, including routing, information brokerage, service establishement, sensor tasking and control, and distributed data storage. A special challenge is the integration of techniques from a variety of disciplines that comes into play in supporting high-level sensor netowrk information management - including signal processing, networking, energy-aware computing, distributed databases and algorithms, and embedded systems and platforms.