Browse Prior Art Database

A METHOD OF DISPLAYING HUGE DATASETS USING ZOOMING LEVELS

IP.com Disclosure Number: IPCOM000222528D
Publication Date: 2012-Oct-16
Document File: 4 page(s) / 155K

Publishing Venue

The IP.com Prior Art Database

Abstract

When displaying datasets on a device such as an oscilloscope or a spectrum analyzer, some form of decimation is typically used to reduce the number of samples of a dataset stored within display memory of the device. In the case that the samples are to be displayed in real-time, the samples are decimated when they are received by the device, before display. Decimation is done intelligently, where the user may select from among three different modes such as for example a sample mode in which the first sample of a data block is selected for display, a maximum mode in which a largest sample of a data block is selected for display, and an average mode in which an average of all the samples of a data block is calculated and displayed.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 33% of the total text.

A method of Displaying Huge Datasets Using Zooming Levels

Jurgen Vandermot

Gery Verhaegen

Joeri Melis

Background

When displaying datasets on a device such as an oscilloscope or a spectrum analyzer, some form of decimation is typically used to reduce the number of samples of a dataset stored within display memory of the device.  In the case that the samples are to be displayed in real-time, the samples are decimated when they are received by the device, before display.  Decimation is done intelligently, where the user may select from among three different modes such as for example a sample mode in which the first sample of a data block is selected for display, a maximum mode in which a largest sample of a data block is selected for display, and an average mode in which an average of all the samples of a data block is calculated and displayed.

In some equipment, sample values are stored on a hard disk or similar source, and loaded from the disk to a display memory within the device.  This makes it possible to analyze huge datasets.  However, a problem with displaying huge datasets of giga or tera sample size is that the display memory of the device is too small.  Decimating such a huge dataset before loading it into a display memory may undesirably remove important information, preventing display of waveform anomalies.  Also, the time needed to load and display such huge datasets may be unacceptable from a user standpoint.

 

                                                            Description

In accordance with the present disclosure, huge datasets of giga and tera sample size may be displayed without the need of display memory large enough to store a complete dataset, and without the loss of important details of the dataset to be displayed.  The method enables visual inspection and analysis of the entire dataset with certainty, and can be used to display waveforms. 

Initially, a detailed visualization where all samples are displayed is only needed when the number of samples is equal to or even smaller than the resolution of the display width.  For huge datasets of giga and tera sample size, this only occurs when the user maximally zooms in to look at a small part of the dataset.  Upon zooming out to a certain level, several samples must be fit onto a single pixel, and consequently details may be lost during visualization even if all samples of the dataset were sent to display memory of the device.

For example, Fig. 1A shows a waveform having 1,000 samples on a display having a display width of 1,000 pixels.  Fig. 1B shows the same waveform zoomed out by 25x, with 25,000 samples on a display having the same 1,000 pixel display width.  The original set of data samples are decimated so that each of the 1,000 pixels of the display represent 25 samples.  At the zooming level shown in Fig. 1B, an intelligently decimated set of samples can be generated and sent to the display...