People | Publications

---

Projects | Funding

---

Looking forward 10-20 years we envision Internet scale sensing where the majority of the traffic on the network is sensor data and the majority of the applications used every day by people integrate sensing, fusion, and actuation in some form or another.

MetroSense is a new wireless sensor edge network for the future Internet based on the concept of "people-centric sensing" at scale (e.g., campus, town, metropolis).

When we think of existing sensor networks, people are out-of-the-loop they simply interact at the periphery of the network with physically embedded static sensor webs to realize small scale application-specific sensing applications, such as environmental and industrial monitoring. MetroSense juxtaposes this view of sensing with that of the traditional view of wireless sensor networking; not only are people in-the-loop but they are central to the sensing experience and represent the key architectural component of the new sensor edge. MetroSense transforms sensing from mostly static and physically embedded to mostly mobile and people-centric. Sensing is no longer application specific but provides an open general-purpose programming environment capable of the execution of multiple distributed sensing applications in parallel.

We are studying three aspects of the MetroSense architecture: the large-scale deployment of mobile people-centric sensors (both motes and sensor-enabled cell phones) and their interaction with embedded static sensor webs; the concept of opportunistic tasking, sensing, and fusion; and security, trust, and privacy - because people-centric sensing raises a number of important privacy issues.

To evaluate MetroSense we are building a large-scale campus-area sensor network at Dartmouth College and ultimately across parts of Hanover with the goal of enabling people-centric applications based on secure, scalable sensing and fusion.

MetroSense presents a number of new and interesting challenges that ultimately might help shape sensing as it shifts from its small-scale application-specific mutli-hop origins to a new sensor edge for Internet - where sensing is no longer a niche application but a part of the mainstream of people's every day lives.

The MetroSense Project is a collaborative project between Dartmouth College's ISTS, Computer Science, and the Thayer School of Engineering, Columbia University's Electrical Engineering, and Intel Corporation and Nokia Research.