Challenges in Programming Sensor Networks

Feng Zhao, Microsoft Research

The proliferation of networked embedded devices such as wireless sensors ushers in an entirely new class of computing platforms. We need new ways to organize and program them. Unlike existing platforms, systems such as sensor networks are decentralized, embedded in physical world, and interact with people. In addition to computing, energy and bandwidth resources are constrained and must be negotiated. Uncertainty, both in systems and about the environment, is a given. Many tasks require collaboration among devices, and the entire network may have to be regarded as a processor.

We argue that the existing node-centric programming of embedded devices is inadequate and unable to scale up. We need new service architectures, inter-operation protocols, programming models that are resource-aware and resource-efficient across heterogeneous devices that can range from extremely limited sensor motes to more powerful servers. I will supplement these discussions with concrete examples arising from our own work and the work of others.

Speaker Biography

Feng Zhao is a Senior Researcher at Microsoft, where he manages the Networked Embedded Computing Group. He received his PhD in Electrical Engineering and Computer Science from MIT and has taught at Stanford University and Ohio State University. Dr. Zhao was a Principal Scientist at Xerox PARC and founded PARC?s sensor network research effort. He serves as the founding Editor-In-Chief of ACM Transactions on Sensor Networks, and has authored or co-authored more than 100 technical papers and books, including a recent book published by Morgan Kaufmann - Wireless Sensor Networks: An information processing approach. He has received a number of awards, and his work has been featured in news media such as BBC World News, BusinessWeek, and Technology Review.