Browse Prior Art Database

Repeatable, Robust Level Measurement Algorithm for Digitally Sampled Periodic Waveforms

IP.com Disclosure Number: IPCOM000034166D
Original Publication Date: 1989-Jan-01
Included in the Prior Art Database: 2005-Jan-26
Document File: 2 page(s) / 33K

Publishing Venue

IBM

Related People

Harrington, RJ: AUTHOR [+4]

Abstract

This article describes an algorithm for making meaningful, repeatable level measurements on digitally sampled periodic waveforms. The up and down level are one of the most important characteristics of a periodic waveform. They are useful measurements by themselves, but they are also necessary for making several other measurements of interest, including the rise and fall times, period and delay. The most commonly used algorithm in digital scopes is a histogramming technique. This algorithm is simple to implement but is not meaningful and is also unstable for many "real life" waveforms. The algorithm presented here essentially averages all the data points between the corners (start and end of the up and down level regions as defined below) to arrive at an up and down level.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 64% of the total text.

Page 1 of 2

Repeatable, Robust Level Measurement Algorithm for Digitally Sampled Periodic Waveforms

This article describes an algorithm for making meaningful, repeatable level measurements on digitally sampled periodic waveforms. The up and down level are one of the most important characteristics of a periodic waveform. They are useful measurements by themselves, but they are also necessary for making several other measurements of interest, including the rise and fall times, period and delay. The most commonly used algorithm in digital scopes is a histogramming technique. This algorithm is simple to implement but is not meaningful and is also unstable for many "real life" waveforms. The algorithm presented here essentially averages all the data points between the corners (start and end of the up and down level regions as defined below) to arrive at an up and down level.

The most difficult task is in finding the corners in a quick, non-iterative manner. The algorithm proceeds as follows: 1. The discrete derivative is calculated as [(Y(N+1)-Y(N-1))/2]/DX, where Y(N) is the Nth sample of

the acquired waveform, and DX is the relative distance

between points (i.e. 1/total number of points). Units

and scaling are unimportant since the algorithm looks

for relative changes. Other discrete derivative

algorithms can also be used. 2. To increase repeatability and noise immunity, the derivative is smoothed. One easy way to accomplish

this is to convolve the derivative with a square wave...