Designing High-Performance Green Filters Using Downsampling Techniques

David Shiung, Wen Long Chin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Reduction of the power consumed by highperformance filters is the ultimate objective of all digital signal processing (DSP) practitioners. However, low power consumption and high-performance are often contradictory design goals. This paper uses a technique involves extending the applicability of the Noble identity in commuting downsamplers with 'sparse transfer functions,' which can be expressed as a function of ZM, where M = 2,3,4,.... More precisely, the transfer functions are relaxed so that they can be expressed as a function of Z1. This commuting relation, termed the generalized Noble identity (GNI), substantially reduces the power consumption of a filter. The GNI is applied to elaborately designed lowpass filters (LPFs) and achieves a favorable compromise between contradictory design goals.

Original languageEnglish
Title of host publicationProceedings - 2019 International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728130385
DOIs
Publication statusPublished - 2019 Dec
Event2019 International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2019 - Taipei, Taiwan
Duration: 2019 Dec 32019 Dec 6

Publication series

NameProceedings - 2019 International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2019

Conference

Conference2019 International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2019
CountryTaiwan
CityTaipei
Period19-12-0319-12-06

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'Designing High-Performance Green Filters Using Downsampling Techniques'. Together they form a unique fingerprint.

Cite this