Accession Number : ADA300902
Title : High-Level Adaptive Signal Processing Architecture with Applications to Radar Non-Gaussian Clutter. Volume 2. A New Technique for Distribution Approximation of Random Data.
Descriptive Note : Final rept. Apr 91-Jun 94,
Corporate Author : MASSACHUSETTS UNIV AMHERST DEPT OF COMPUTER SCIENCE
Personal Author(s) : Shah, Rajiv R.
PDF Url : ADA300902
Report Date : SEP 1995
Pagination or Media Count : 100
Abstract : This thesis deals with the analysis of random data. Two approaches are discussed. The first approach is a Goodness of Fit test to determine whether or not random data samples are statistically consistent with a prespecified probability distribution. The well known Kolmogorov Smirnov test, Chi Square test, Q-Q Plots and P-P plots are reviewed and illustrated by means of several examples. A new algorithm, the Ozturk Algorithm, is introduced. The second approach deals with approximation of the underlying probability density function of random data samples. The previously mentioned well known tests are not suitable for this task. However, the Ozturk Algorithm provides a powerful solution for this problem with a nice graphical interpretation. Finally, computer simulated results obtained with the Ozturk Algorithm are presented and discussed.
Descriptors : *SIGNAL PROCESSING, *RADAR SIGNALS, ALGORITHMS, COMPUTERIZED SIMULATION, DATA PROCESSING, COMPUTER ARCHITECTURE, PROBABILITY DENSITY FUNCTIONS, SAMPLING, RADAR CLUTTER, CHI SQUARE TEST.
Subject Categories : Active & Passive Radar Detection & Equipment
Distribution Statement : APPROVED FOR PUBLIC RELEASE