Independent Randomness Tests based on the Orthogonalized Non-overlapping Template Matching Test

08/20/2019
by   Atsushi Iwasaki, et al.
0

In general, randomness tests included in a test suite are not independent of each other. This renders it difficult to fix a rational criterion through the whole test suite with an explicit significance level. In this paper, we focus on the Non-overlapping Template Matching Test, which is a randomness test included in the NIST statistical test suite. The test uses a parameter called "template" and we can consider a test item for each template. We investigate dependency between two test items by deriving the joint probability density function of the two p-values and propose a transformation to make multi test items independent of each other.

READ FULL TEXT

page 7

page 9

research
10/04/2017

Analysis of NIST SP800-22 focusing on randomness of each sequence

NIST SP800-22 is a randomness test set applied for a set of sequences. A...
research
04/11/2011

Template-based matching using weight maps

Template matching is one of the most prevalent pattern recognition metho...
research
06/03/2019

The performance of the partially overlapping samples t-tests at the limits

This note includes an assessment of the partially overlapping samples t-...
research
10/15/2021

Second-level randomness test based on the Kolmogorov-Smirnov test

We analyzed the effect of the deviation of the exact distribution of the...
research
08/26/2023

The Spearman-Brown Formula and Reliabilities of Random Test Forms

It is shown that the psychometric test reliability, based on any true-sc...
research
09/10/2023

Testing for Stationary or Persistent Coefficient Randomness in Predictive Regressions

This study considers tests for coefficient randomness in predictive regr...
research
09/23/2022

A Neural Template Matching Method to Detect Knee Joint Areas

In this paper, new methods are considered to detect knee joint areas in ...

Please sign up or login with your details

Forgot password? Click here to reset