Browse Prior Art Database

Parallel Adaptive Composite Data Compressor

IP.com Disclosure Number: IPCOM000093573D
Original Publication Date: 1967-Nov-01
Included in the Prior Art Database: 2005-Mar-06
Document File: 2 page(s) / 29K

Publishing Venue

IBM

Related People

Schwartz, GR: AUTHOR [+2]

Abstract

A number of algorithms are available which have the ability to predict future data points given a set of data points. These algorithms can be used in order to compress a set of data which is to be transmitted over a communication network. In drawing 1 a graph is shown of some varying quantity, e.g., temperature. The latter is sampled at discrete time intervals T, resulting in a given set of data a...m. Applying one of these prediction algorithms, point d might be predicted if points a, b, and c are given. Moreover, points e and f might also be predicted, but not point g. Therefore, utilizing this algorithm in a communication network, it would be necessary only to transmit points a, b, c, and g.

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

Parallel Adaptive Composite Data Compressor

A number of algorithms are available which have the ability to predict future data points given a set of data points. These algorithms can be used in order to compress a set of data which is to be transmitted over a communication network. In drawing 1 a graph is shown of some varying quantity, e.g., temperature. The latter is sampled at discrete time intervals T, resulting in a given set of data a...m. Applying one of these prediction algorithms, point d might be predicted if points a, b, and c are given. Moreover, points e and f might also be predicted, but not point g. Therefore, utilizing this algorithm in a communication network, it would be necessary only to transmit points a, b, c, and g.

However, no algorithm is suitable for all circumstances. One algorithm is superior for predicting data values that are quickly changing. Another algorithm is superior for predicting data values that are constant. In order to take advantage of this characteristic, a number of algorithms are utilized at once. In the example shown in drawing 2, only two algorithms are used. However, this device can be extended to three or more algorithms.

The signal representing the sample data point is fed simultaneously into computer 1, computer 2, and comparator 3. Computers 1 and 2 compute the next data point given the ones that they have already received. Each computer contains a different algorithm. Comparator 3 compares the predicted data poi...