Share Email Print

Proceedings Paper

Cost function design for modeling information overload in radar systems
Author(s): Paul G. Singerman; Ram M. Narayanan; Muralidhar Rangaswamy
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

It is a common assumption that more effective decisions can be made in radar sensor processing if more “information" is available. However, this is not always the case since more information may not necessarily translate to better sensor performance due to a phenomenon known as information overload. What determines the effectiveness of a decision is rarely only dependent on a single performance metric. Instead, there are often multiple performance metrics involved. This necessitates the use of cost functions for modeling a loss in decision effectiveness due to deviations from the optimal values of the performance metrics. These optimal values are also often determined in a fuzzy/subjective manner by an individual using language. Two different types of statements are looked at and modeled, namely, soft and hard constraining statements. These models, called language-based cost functions (LBCFs), are derived and their properties are explored. These can then be easily combined, via a summation, to obtain a smooth objective function. The objective function is then used to find the utility of the radar system. By altering the systems parameter values the amount of information introduced into the system can be varied and the point at which the utility of the radar system decreases due to excess information can be found. This is then regarded as the information overload point or the point at which more information causes a decrease in the utility of the system. An example of a radar system with imposed constraints is provided and its resulting information overload point is found by utilizing LBCFs.

Paper Details

Date Published: 3 May 2019
PDF: 11 pages
Proc. SPIE 11003, Radar Sensor Technology XXIII, 110030A (3 May 2019); doi: 10.1117/12.2519726
Show Author Affiliations
Paul G. Singerman, The Pennsylvania State Univ. (United States)
Ram M. Narayanan, The Pennsylvania State Univ. (United States)
Muralidhar Rangaswamy, Air Force Research Lab. (United States)

Published in SPIE Proceedings Vol. 11003:
Radar Sensor Technology XXIII
Kenneth I. Ranney; Armin Doerry, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?