Shimokawa, Hidetoshi (ERATO, JST), Aihara, Kazuyuki (The Univ. of Tokyo)

State Estimation under Information Constraints

Scheduled for presentation during the Mini-Symposium "Data Rate Constraints in Networked Control and Estimation" (FrP12), Friday, July 28, 2006, 15:20−15:45, Room 104b

17th International Symposium on Mathematical Theory of Networks and Systems, July 24-28, 2006, Kyoto, Japan

This information is tentative and subject to change. Compiled on April 24, 2024

Abstract

Many sensor networks carry quantized data these days and it is interesting problem to know how much data rate is necessary and how we can encode efficiently for estimating state of a linear system. Preceding study shows that if full information is available at the encoder, we can estimate the state in arbitrary precision asymptotically with finite data rate which is determined by instability of the plant. In this paper, we propose a new quantization method that achieves near-optimum data rate even if the encoder is memory-less and does not know inputs to the plant. This result is related to so-called side information problem in information theory. We can consider inputs to the plant as side information available only at the decoder.