|
Abstract:
|
Statistically more test samples obtained from a single chip would give a better picture of the various noise processes present . Increasing the number of samples while testing one chip would however lead to an increase in the testing time , decreasing the overall throughput . The aim of this report is to investigate the detection of non -Gaussian noise (burst noise ) in a random set of data with a small number of samples .
In order to determine whether a given set of noise samples has non -Gaussian noise processes present , a Chi -Squared ‘Goodness of Fit’ test on a modeled set of random data is presented . A discussion of test methodologies using a single test measurement pass as well as two passes is presented from the obtained simulation results . |