[Colloq] Hiring Talk - Tuesday, March 10 - Anand Panangdan

[Rachel Kalweit]

The College of Computer and Information Science presents a hiring talk by:
Anand Panangdan - Saban Research Institute of the Childrens Hospital, Los Angeles

Tuesday, March 10, 2009 at 1:30pm
Location: 366 West Village H

Title: Resource Management using Adaptive Sensing for a Body Area Network

Abstract:
Recent advances in sensor and communication technology combined with better signal processing techniques have made sensor networks valuable in applications such as environment monitoring. These advances also make wireless sensor networks suitable for long-term human health monitoring. A ?body area network? is a wireless network of biomedical sensors that are attached to a human subject's body. The goal is to develop a system that will enable a person's health condition to be continuously recorded and monitored if needed over a long distance communication network. As the sensing system is to be worn by a subject for long durations, the hardware needs to be compact and light. In particular, this limits the size of the batteries. These factors have made energy the most critical resource in body area networks and extending system lifetime has become a priority to fully realize the capabilities of body area networks.
In this talk, I describe a body area network developed at the Saban Research Institute of the Childrens Hospital Los Angeles. The network incorporates noninvasive physiological sensors (such as pulse oximeters) and a novel minimally invasive metabolite sensor. I will describe the energy conservation algorithms that are implemented at multiple levels of the system. These algorithms adapt the operating parameters of the sensors such as sampling rate to the time-varying system resources and criticality of the measured data (adaptive sensing). I have used the theory of Markov Decision Processes in the design of these algorithms. The adaptive sensing policy is a joint policy ? it takes into account the state of all the sensors to determine the optimum sampling rates. A na?ve implementation of this policy will thus require the continuous exchange of state information between the distributed sensors. I will describe an information theoretic method for quantifying the expected value of communication. Sensors will communicate their state only when this value exceeds some threshold.

Bio:
Anand Panangadan is a Research Specialist at the Saban Research Institute of the Childrens Hospital Los Angeles and a post-doctoral affiliate at NASA?s Jet Propulsion Laboratory. His current research is on developing information processing and resource management algorithms for wireless sensor networks. These algorithms are deployed in human health monitoring systems (?body area networks?) and environmental sensor networks. He received the Ph.D. degree in Computer Science from the University of California, Los Angeles in 2002 and the B.Tech. degree in Computer Science and Engineering from the Indian Institute of Technology, Bombay in 1996. He was a post-doctoral researcher at the USC Robotics Research Labs during 2003-2004.
Webpage: http://www-hsc.usc.edu/~anandvp

Host: Tim Bickmore