Information-driven sensor planning: Navigating a statistical manifold

Douglas Cochran, Alfred O. Hero

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Many adaptive sensing and sensor management strategies seek to determine a sequence of sensor actions that successively optimizes an objective function. Frequently the goal is to adjust a sensor to best estimate a partially observed state variable, for example, the objective function may be the final mean-squared state estimation error. Information-driven sensor planning strategies adopt an objective function that measures the accumulation of information as defined by a suitable metric, such as Fisher information, Bhattacharyya affinity, or Kullback-Leibler divergence. These information measures are defined on the space of probability distributions of data acquired by the sensor, and there is a distribution in this space corresponding to each sensor configuration. Hence, sensor planning can be posed as a problem of optimally navigating over a statistical manifold of probability distributions. This information-geometric perspective presents new insights into adaptive sensing and sensor management.

Original languageEnglish (US)
Title of host publication2013 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2013 - Proceedings
Pages1049-1052
Number of pages4
DOIs
StatePublished - 2013
Event2013 1st IEEE Global Conference on Signal and Information Processing, GlobalSIP 2013 - Austin, TX, United States
Duration: Dec 3 2013Dec 5 2013

Other

Other2013 1st IEEE Global Conference on Signal and Information Processing, GlobalSIP 2013
CountryUnited States
CityAustin, TX
Period12/3/1312/5/13

Fingerprint

Planning
Sensors
Probability distributions
State estimation

Keywords

  • Adaptive sensing
  • Hellinger distance
  • Information geometry
  • Multinomial class of distributions
  • Sensor management

ASJC Scopus subject areas

  • Information Systems
  • Signal Processing

Cite this

Cochran, D., & Hero, A. O. (2013). Information-driven sensor planning: Navigating a statistical manifold. In 2013 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2013 - Proceedings (pp. 1049-1052). [6737074] https://doi.org/10.1109/GlobalSIP.2013.6737074

Information-driven sensor planning : Navigating a statistical manifold. / Cochran, Douglas; Hero, Alfred O.

2013 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2013 - Proceedings. 2013. p. 1049-1052 6737074.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Cochran, D & Hero, AO 2013, Information-driven sensor planning: Navigating a statistical manifold. in 2013 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2013 - Proceedings., 6737074, pp. 1049-1052, 2013 1st IEEE Global Conference on Signal and Information Processing, GlobalSIP 2013, Austin, TX, United States, 12/3/13. https://doi.org/10.1109/GlobalSIP.2013.6737074
Cochran D, Hero AO. Information-driven sensor planning: Navigating a statistical manifold. In 2013 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2013 - Proceedings. 2013. p. 1049-1052. 6737074 https://doi.org/10.1109/GlobalSIP.2013.6737074
Cochran, Douglas ; Hero, Alfred O. / Information-driven sensor planning : Navigating a statistical manifold. 2013 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2013 - Proceedings. 2013. pp. 1049-1052
@inproceedings{cd794256c2db4d6fb00a3993e9e67ef3,
title = "Information-driven sensor planning: Navigating a statistical manifold",
abstract = "Many adaptive sensing and sensor management strategies seek to determine a sequence of sensor actions that successively optimizes an objective function. Frequently the goal is to adjust a sensor to best estimate a partially observed state variable, for example, the objective function may be the final mean-squared state estimation error. Information-driven sensor planning strategies adopt an objective function that measures the accumulation of information as defined by a suitable metric, such as Fisher information, Bhattacharyya affinity, or Kullback-Leibler divergence. These information measures are defined on the space of probability distributions of data acquired by the sensor, and there is a distribution in this space corresponding to each sensor configuration. Hence, sensor planning can be posed as a problem of optimally navigating over a statistical manifold of probability distributions. This information-geometric perspective presents new insights into adaptive sensing and sensor management.",
keywords = "Adaptive sensing, Hellinger distance, Information geometry, Multinomial class of distributions, Sensor management",
author = "Douglas Cochran and Hero, {Alfred O.}",
year = "2013",
doi = "10.1109/GlobalSIP.2013.6737074",
language = "English (US)",
isbn = "9781479902484",
pages = "1049--1052",
booktitle = "2013 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2013 - Proceedings",

}

TY - GEN

T1 - Information-driven sensor planning

T2 - Navigating a statistical manifold

AU - Cochran, Douglas

AU - Hero, Alfred O.

PY - 2013

Y1 - 2013

N2 - Many adaptive sensing and sensor management strategies seek to determine a sequence of sensor actions that successively optimizes an objective function. Frequently the goal is to adjust a sensor to best estimate a partially observed state variable, for example, the objective function may be the final mean-squared state estimation error. Information-driven sensor planning strategies adopt an objective function that measures the accumulation of information as defined by a suitable metric, such as Fisher information, Bhattacharyya affinity, or Kullback-Leibler divergence. These information measures are defined on the space of probability distributions of data acquired by the sensor, and there is a distribution in this space corresponding to each sensor configuration. Hence, sensor planning can be posed as a problem of optimally navigating over a statistical manifold of probability distributions. This information-geometric perspective presents new insights into adaptive sensing and sensor management.

AB - Many adaptive sensing and sensor management strategies seek to determine a sequence of sensor actions that successively optimizes an objective function. Frequently the goal is to adjust a sensor to best estimate a partially observed state variable, for example, the objective function may be the final mean-squared state estimation error. Information-driven sensor planning strategies adopt an objective function that measures the accumulation of information as defined by a suitable metric, such as Fisher information, Bhattacharyya affinity, or Kullback-Leibler divergence. These information measures are defined on the space of probability distributions of data acquired by the sensor, and there is a distribution in this space corresponding to each sensor configuration. Hence, sensor planning can be posed as a problem of optimally navigating over a statistical manifold of probability distributions. This information-geometric perspective presents new insights into adaptive sensing and sensor management.

KW - Adaptive sensing

KW - Hellinger distance

KW - Information geometry

KW - Multinomial class of distributions

KW - Sensor management

UR - http://www.scopus.com/inward/record.url?scp=84897699169&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897699169&partnerID=8YFLogxK

U2 - 10.1109/GlobalSIP.2013.6737074

DO - 10.1109/GlobalSIP.2013.6737074

M3 - Conference contribution

AN - SCOPUS:84897699169

SN - 9781479902484

SP - 1049

EP - 1052

BT - 2013 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2013 - Proceedings

ER -