We examine the problem of tracking objects using a sensor network. In order to conserve energy, the sensors are allowed to enter an inactive state at the expense of observability. We focus on two variants of the sensor management problem, namely, sensor scheduling and sensor sleeping. The former refers to the scenario where the sensors can be turned on or off at consecutive time steps, while the latter refers to the scenario where an asleep sensor cannot be communicated with or woken up, and hence the sleep duration needs to be determined at the time the sensor goes to sleep based on all the information available to the sensor. Our objective is to find control policies to optimize the tradeoff between tracking performance and energy consumption for these scenarios.
We study a number of different formulations of these problems, starting with simple models for object movement and sensing and ending with more complicated and realistic models. While optimal solutions to these problems are generally intractable, we show that we can find good control policies that approach computable lower bounds on optimal performance. More importantly, these control policies significantly outperform traditional approaches based on duty cycling the sensors between on and off states.
George Atia joined the Department of Electrical and Computer Engineering at the University of Illinois at Urbana-Champaign in Fall 2009, where he is currently a postdoctoral research associate with the Coordinated Science Laboratory. He received his Ph.D. degree in Electrical and Computer Engineering from Boston University, Massachusetts, in 2009. He received the B.Sc. and M.Sc. degrees, both in Electrical Engineering, from Alexandria University, Egypt, in 2000 and 2003, respectively. He is the recipient of many awards including the outstanding graduate teaching fellow of the year award in 2003-2004, the 2006 College of Engineering Deans Award at the Science and Engineering Research Symposium, and the best paper award at the International Conference on Distributed Computing in Sensor Systems (DCOSS) in 2008. His main research interests are in wireless communications, statistical signal processing and information theory. His current research focus is on controlled sensing for inference and sparse signal processing.
