A coarse grained and hybrid reconfigurable architecture with flexible NOC router for variable block size motion estimation

Ruchika Verma, Ali Akoglu

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

3 Citations (Scopus)

Abstract

This paper proposes a novel application-specific hybrid coarse-grained reconfigurable architecture with a flexible network on chip (NoC) mechanism. Architecture supports variable block size motion estimation (VBSME) with much less resources than ASIC based and coarse grained reconfigurable architectures. The intelligent NoC router supports full search motion estimation algorithm as well as other fast search algorithms like diamond, hexagon, big hexagon and spiral. Our model is a hierarchical hybrid processing element based 2D architecture which supports reuse of reference frame blocks between the processing elements through NoC routers. This reduces the transactions from/to the main memory. Proposed architecture is designed with Verilog-HDL description and synthesized by 90 nm CMOS standard cell library. Results show that our architecture reduces the gate count by 7x compared to its ASIC counterpart that only supports full search method. Moreover, the proposed architecture operates at a frequency comparable to ASIC based implementation to sustain 30fps. Our approach is based on a simple design which utilizes a high-level of parallelism with an intensive data reuse. Therefore, proposed architecture supports run-time reconfiguration for any block size and for any search pattern depending on the application requirement.

Original languageEnglish (US)
Title of host publicationIPDPS Miami 2008 - Proceedings of the 22nd IEEE International Parallel and Distributed Processing Symposium, Program and CD-ROM
DOIs
StatePublished - 2008
EventIPDPS 2008 - 22nd IEEE International Parallel and Distributed Processing Symposium - Miami, FL, United States
Duration: Apr 14 2008Apr 18 2008

Other

OtherIPDPS 2008 - 22nd IEEE International Parallel and Distributed Processing Symposium
CountryUnited States
CityMiami, FL
Period4/14/084/18/08

Fingerprint

Reconfigurable architectures
Motion estimation
Application specific integrated circuits
Routers
Intelligent networks
Computer hardware description languages
Processing
Diamonds
Data storage equipment
Network-on-chip

ASJC Scopus subject areas

  • Hardware and Architecture
  • Software
  • Electrical and Electronic Engineering

Cite this

Verma, R., & Akoglu, A. (2008). A coarse grained and hybrid reconfigurable architecture with flexible NOC router for variable block size motion estimation. In IPDPS Miami 2008 - Proceedings of the 22nd IEEE International Parallel and Distributed Processing Symposium, Program and CD-ROM [4536528] https://doi.org/10.1109/IPDPS.2008.4536528

A coarse grained and hybrid reconfigurable architecture with flexible NOC router for variable block size motion estimation. / Verma, Ruchika; Akoglu, Ali.

IPDPS Miami 2008 - Proceedings of the 22nd IEEE International Parallel and Distributed Processing Symposium, Program and CD-ROM. 2008. 4536528.

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

Verma, R & Akoglu, A 2008, A coarse grained and hybrid reconfigurable architecture with flexible NOC router for variable block size motion estimation. in IPDPS Miami 2008 - Proceedings of the 22nd IEEE International Parallel and Distributed Processing Symposium, Program and CD-ROM., 4536528, IPDPS 2008 - 22nd IEEE International Parallel and Distributed Processing Symposium, Miami, FL, United States, 4/14/08. https://doi.org/10.1109/IPDPS.2008.4536528
Verma R, Akoglu A. A coarse grained and hybrid reconfigurable architecture with flexible NOC router for variable block size motion estimation. In IPDPS Miami 2008 - Proceedings of the 22nd IEEE International Parallel and Distributed Processing Symposium, Program and CD-ROM. 2008. 4536528 https://doi.org/10.1109/IPDPS.2008.4536528
Verma, Ruchika ; Akoglu, Ali. / A coarse grained and hybrid reconfigurable architecture with flexible NOC router for variable block size motion estimation. IPDPS Miami 2008 - Proceedings of the 22nd IEEE International Parallel and Distributed Processing Symposium, Program and CD-ROM. 2008.
@inproceedings{b76db88869164e54844a2d7ff0576281,
title = "A coarse grained and hybrid reconfigurable architecture with flexible NOC router for variable block size motion estimation",
abstract = "This paper proposes a novel application-specific hybrid coarse-grained reconfigurable architecture with a flexible network on chip (NoC) mechanism. Architecture supports variable block size motion estimation (VBSME) with much less resources than ASIC based and coarse grained reconfigurable architectures. The intelligent NoC router supports full search motion estimation algorithm as well as other fast search algorithms like diamond, hexagon, big hexagon and spiral. Our model is a hierarchical hybrid processing element based 2D architecture which supports reuse of reference frame blocks between the processing elements through NoC routers. This reduces the transactions from/to the main memory. Proposed architecture is designed with Verilog-HDL description and synthesized by 90 nm CMOS standard cell library. Results show that our architecture reduces the gate count by 7x compared to its ASIC counterpart that only supports full search method. Moreover, the proposed architecture operates at a frequency comparable to ASIC based implementation to sustain 30fps. Our approach is based on a simple design which utilizes a high-level of parallelism with an intensive data reuse. Therefore, proposed architecture supports run-time reconfiguration for any block size and for any search pattern depending on the application requirement.",
author = "Ruchika Verma and Ali Akoglu",
year = "2008",
doi = "10.1109/IPDPS.2008.4536528",
language = "English (US)",
isbn = "9781424416943",
booktitle = "IPDPS Miami 2008 - Proceedings of the 22nd IEEE International Parallel and Distributed Processing Symposium, Program and CD-ROM",

}

TY - GEN

T1 - A coarse grained and hybrid reconfigurable architecture with flexible NOC router for variable block size motion estimation

AU - Verma, Ruchika

AU - Akoglu, Ali

PY - 2008

Y1 - 2008

N2 - This paper proposes a novel application-specific hybrid coarse-grained reconfigurable architecture with a flexible network on chip (NoC) mechanism. Architecture supports variable block size motion estimation (VBSME) with much less resources than ASIC based and coarse grained reconfigurable architectures. The intelligent NoC router supports full search motion estimation algorithm as well as other fast search algorithms like diamond, hexagon, big hexagon and spiral. Our model is a hierarchical hybrid processing element based 2D architecture which supports reuse of reference frame blocks between the processing elements through NoC routers. This reduces the transactions from/to the main memory. Proposed architecture is designed with Verilog-HDL description and synthesized by 90 nm CMOS standard cell library. Results show that our architecture reduces the gate count by 7x compared to its ASIC counterpart that only supports full search method. Moreover, the proposed architecture operates at a frequency comparable to ASIC based implementation to sustain 30fps. Our approach is based on a simple design which utilizes a high-level of parallelism with an intensive data reuse. Therefore, proposed architecture supports run-time reconfiguration for any block size and for any search pattern depending on the application requirement.

AB - This paper proposes a novel application-specific hybrid coarse-grained reconfigurable architecture with a flexible network on chip (NoC) mechanism. Architecture supports variable block size motion estimation (VBSME) with much less resources than ASIC based and coarse grained reconfigurable architectures. The intelligent NoC router supports full search motion estimation algorithm as well as other fast search algorithms like diamond, hexagon, big hexagon and spiral. Our model is a hierarchical hybrid processing element based 2D architecture which supports reuse of reference frame blocks between the processing elements through NoC routers. This reduces the transactions from/to the main memory. Proposed architecture is designed with Verilog-HDL description and synthesized by 90 nm CMOS standard cell library. Results show that our architecture reduces the gate count by 7x compared to its ASIC counterpart that only supports full search method. Moreover, the proposed architecture operates at a frequency comparable to ASIC based implementation to sustain 30fps. Our approach is based on a simple design which utilizes a high-level of parallelism with an intensive data reuse. Therefore, proposed architecture supports run-time reconfiguration for any block size and for any search pattern depending on the application requirement.

UR - http://www.scopus.com/inward/record.url?scp=51049118363&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51049118363&partnerID=8YFLogxK

U2 - 10.1109/IPDPS.2008.4536528

DO - 10.1109/IPDPS.2008.4536528

M3 - Conference contribution

AN - SCOPUS:51049118363

SN - 9781424416943

BT - IPDPS Miami 2008 - Proceedings of the 22nd IEEE International Parallel and Distributed Processing Symposium, Program and CD-ROM

ER -