Share Email Print

Proceedings Paper

Adaptive Grahm-Schmidt orthogonalization for the projection-slice synthetic discriminant function filter
Author(s): Vahid R. Riasati; Denis Grishin
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

One of the primary reasons for using the PSDF is the inherent data dimension reduction that is achieved through the use of the projection-slice theorem, (PST). The use of the PST allows for a powerful technique to segment informational content into lower dimensional spans while simultaneously providing a complete and naturally linked data set. Shannon provides a formula for maximum information capacity in a channel that utilizes as the most efficient informational coding independent data samples, as this spreads the spectrum evenly across the channel band. This implies that independent information that represent a correlated set contains maximal information if the "key" that represents the correlated-ness of the data is known, otherwise the independent data are purely random. By using a novel Adaptive Grahm-Schmidt (AGS) procedure we form a method that identifies patterns in correlated data for the removal of the inter-dependence and thereby maximization of information content per data sample. In this work we subject the lower-dimensional data sets in the PSDF to the AGS to maximize the information content of the PSDF and share some of our findings, results, and deductions.

Paper Details

Date Published: 9 April 2007
PDF: 11 pages
Proc. SPIE 6570, Data Mining, Intrusion Detection, Information Assurance, and Data Networks Security 2007, 65700J (9 April 2007); doi: 10.1117/12.720891
Show Author Affiliations
Vahid R. Riasati, Boeing Satellite Systems (United States)
Denis Grishin, Science Applications International Corp. (United States)

Published in SPIE Proceedings Vol. 6570:
Data Mining, Intrusion Detection, Information Assurance, and Data Networks Security 2007
Belur V. Dasarathy, Editor(s)

© SPIE. Terms of Use
Back to Top