Fading observation alignment via feedback
Proceedings of the 4th International Symposium on Information Processing in Sensor Networks (IPSN 2005), Los Angeles, CA, April 2005.
Download
Adobe Portable Document Format - [PDF]
Poster - [PDF]
Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.
Abstract
In some remote sensing applications, the functional relationship between the source being observed and the sensor readings may not be known. Because of communication constraints, this uncertainty may result in poor end-to-end distortion. If the sensors have some knowledge of their joint statistics, they may be able to communicate collaboratively to combat the channel noise. A model is proposed for capturing some of the uncertainty in the observation process, called a \textit{fading observation model}. An example with fading observations is analysed. For $M$ sensors with no fading there exists a scheme for which the achievable distortion scales with $M$ as $M^{-1}$, but with fading the distortion does not scale with $M$. In this paper, a one-bit feedback scheme is presented that provides enough information about the joint statistics to achieve scaling rates like $M^{-1/3}$. Additional feedback improves the achievable scaling rate. For comparison, a scheme based on separate source and channel coding at best gives a distortion scaling behaviour of $(\log M)^{-1}$. Some extensions to multiple sources and observation models with unknown delay are discussed.