Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Digital Filter for Delta Demodulation

IP.com Disclosure Number: IPCOM000093368D
Original Publication Date: 1967-Sep-01
Included in the Prior Art Database: 2005-Mar-06
Document File: 2 page(s) / 24K

Publishing Venue

IBM

Related People

Deutsch, H: AUTHOR [+3]

Abstract

This digital filtering method can be used in the reconstitution of an analog signal from a delta modulated digital signal. The characteristics of the filtering method are such as to eliminate unwanted quantizing noise components which are present in the frequency spectrum of the digital signal and appear as noise in the reconstructed analog signal.

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 86% of the total text.

Page 1 of 2

Digital Filter for Delta Demodulation

This digital filtering method can be used in the reconstitution of an analog signal from a delta modulated digital signal. The characteristics of the filtering method are such as to eliminate unwanted quantizing noise components which are present in the frequency spectrum of the digital signal and appear as noise in the reconstructed analog signal.

These noise components are particularly troublesome when the digital frequency is small compared to the highest frequency of the analog signal being encoded. Analog filters with characteristics suitable for removing these noise components are more expensive than the described method and are critically sensitive to component variations. The digital filter method described below utilizes standard digital circuit blocks.

The digital filtering method is implemented by shifting the digital signal through a shift register. The unweighed outputs from each position of the register are applied to an analog integrator. For example, assume the register has two positions. The alternating bit pattern for a zero derivative analog signal causes no variation in the integrated output. This is because the total input to the integrator from the two register positions does not vary with time. Thus, the frequency of the alternating bit patterns which is one half the sampling frequency is removed from the integrator output. Noise frequencies generated by other pulse patterns can be removed by adding mor...