Dismiss
InnovationQ/InnovationQ Plus content will be updated on Sunday, June 25, 10am ET, with new patent and non-patent literature collections. Click here to learn more.
Browse Prior Art Database

COMPUTING ARRANGEMENT FOR DSP USING ItjTERPOLATlON AND DECIMATION TECHNIQUES

IP.com Disclosure Number: IPCOM000009088D
Original Publication Date: 1999-Jun-01
Included in the Prior Art Database: 2002-Aug-07
Document File: 2 page(s) / 75K

Publishing Venue

Motorola

Related People

Itzhak Barak: AUTHOR [+3]

Abstract

This proposal relates to digital signal processing (DSP) on a multiprocessor arrangement which implements finite impulse response (FIR) filters, infinite impulse response (IIR) filter and other filters.

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 55% of the total text.

Page 1 of 2

@ MOTOROLA

Technical Developments

COMPUTING ARRANGEMENT FOR DSP USING ItjTERPOLATlON AND DECIMATION TECHNIQUES

by ltzhak Barak, Leonid Tsukerman and Jacob A. Kirschenbaum

INTRODUCTION

  This proposal relates to digital signal processing (DSP) on a multiprocessor arrangement which implements finite impulse response (FIR) filters, infinite impulse response (IIR) filter and other filters.

PROBLEM DESCRIPTION

  During filtering, input data samples having an input frequency is consecutively multiplied with fil- ter coefficients (taps) and intermediate results are summed up to output data samples at an output fre- quency. Filtering techniques often use interpolation techniques (upsampling) and decimation techniques (downsampling). Decimation does not consider some of the predetermined samples which however remain as redundant signals in a processor and should be disposed of.

SOLUTION

Figure 1 illustrates a computing arrangement 10 which has

(a) interpolator 20 for interpolating samples X with ZEROS to signal X', (b) a plurality of N

processors 12-n (1 to N) each with computing unit 14-n (CU) and a M-word cache memory 16-n receiving and forwarding X' (14-n to 14-(n+l) ), (c) enable unit 22 for selectively forwarding intermedi- ate calculation results from memory 16 and CUs 14 to (d) adder 18 providing output samples Y. Complete memory 16 (16-1 to 16-N) of arrange- ment 10 has M rows in N columns.

  By combining upsampling (interpolator 20) and downsampling (enable unit 22), a...