High-performance memory interface architecture for high-definition video coding application

This paper proposes new memory interface architecture to overcome huge SDRAM bandwidth requirements of video decoders. To improve the memory efficiency, tile based memory access method and pixel cache are adopted. In addition to large data transfer cycle, extra overhead cycles takes great part of th...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Joon-Ho Song, Doo Hyun Kim, Do-Hyung Kim, Shi Hwa Lee
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 3748
container_issue
container_start_page 3745
container_title
container_volume
creator Joon-Ho Song
Doo Hyun Kim
Do-Hyung Kim
Shi Hwa Lee
description This paper proposes new memory interface architecture to overcome huge SDRAM bandwidth requirements of video decoders. To improve the memory efficiency, tile based memory access method and pixel cache are adopted. In addition to large data transfer cycle, extra overhead cycles takes great part of the heavy memory bandwidth requirement. The overhead cycles are caused by two reasons- the one is due to row-activation overhead of SDRAM and the other is due to redundant data transfer for motion compensation. The row-activation overhead is incurred frequently because of the block by block memory access pattern of standard video decoding algorithms. By using the tile based memory access method, the row-activation overhead can be reduced significantly. The redundant data transfer overhead is due to the 6-tap FIR filtering for the motion compensation while H.264/AVC decoding. We reduced the redundant memory transfer overhead by adopting the reference pixel cache, which is tightly coupled with multi-frame reference picture and the memory structure of the tile based memory access method. Experimental results show that the proposed method can decode 1920 × 1080, 30 frame per second H.264/AVC main profile bitstream in real time with only about 90MHz SDRAM clock. Compared to the conventional linear memory access system, the proposed method reduces about 6 times of SDRAM bandwidth. We also expect the proposed method will reduce power consumption because the SDRAM bandwidth requirement and the number of row-activation overhead are significantly reduced.
doi_str_mv 10.1109/ICIP.2010.5653435
format Conference Proceeding
fullrecord <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_5653435</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>5653435</ieee_id><sourcerecordid>5653435</sourcerecordid><originalsourceid>FETCH-LOGICAL-i241t-b3de5dabd5c2a6a51b37b203d029de1b8636f487d1b1d574cc4b01cb0748db233</originalsourceid><addsrcrecordid>eNpVULtOwzAUNS-JUvoBiMU_kOJrXyfOiCqgkSrBABND5cdNa9Q85Aak_j2p6MJ0dJ7DYewOxBxAlA_VonqbSzFSnWuFSp-xWVkYQIlYlCWaczaRykBmNJYX_zyJl2wCWsoMjRHX7Ga__xJi3FIwYZ_LuNlmPaW6S41tPfGGmi4deGyHUbSjYJPfxoH88J2IjzG-PVYC1bGNQ-xa_hMDddx3IbYbbvt-F709Grfsqra7Pc1OOGUfz0_vi2W2en2pFo-rLEqEIXMqkA7WBe2lza0GpwonhQpCloHAmVzlNZoigIOgC_QenQDvRIEmOKnUlN3_7UYiWvcpNjYd1qef1C8VdFl4</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>High-performance memory interface architecture for high-definition video coding application</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Joon-Ho Song ; Doo Hyun Kim ; Do-Hyung Kim ; Shi Hwa Lee</creator><creatorcontrib>Joon-Ho Song ; Doo Hyun Kim ; Do-Hyung Kim ; Shi Hwa Lee</creatorcontrib><description>This paper proposes new memory interface architecture to overcome huge SDRAM bandwidth requirements of video decoders. To improve the memory efficiency, tile based memory access method and pixel cache are adopted. In addition to large data transfer cycle, extra overhead cycles takes great part of the heavy memory bandwidth requirement. The overhead cycles are caused by two reasons- the one is due to row-activation overhead of SDRAM and the other is due to redundant data transfer for motion compensation. The row-activation overhead is incurred frequently because of the block by block memory access pattern of standard video decoding algorithms. By using the tile based memory access method, the row-activation overhead can be reduced significantly. The redundant data transfer overhead is due to the 6-tap FIR filtering for the motion compensation while H.264/AVC decoding. We reduced the redundant memory transfer overhead by adopting the reference pixel cache, which is tightly coupled with multi-frame reference picture and the memory structure of the tile based memory access method. Experimental results show that the proposed method can decode 1920 × 1080, 30 frame per second H.264/AVC main profile bitstream in real time with only about 90MHz SDRAM clock. Compared to the conventional linear memory access system, the proposed method reduces about 6 times of SDRAM bandwidth. We also expect the proposed method will reduce power consumption because the SDRAM bandwidth requirement and the number of row-activation overhead are significantly reduced.</description><identifier>ISSN: 1522-4880</identifier><identifier>ISBN: 9781424479924</identifier><identifier>ISBN: 1424479924</identifier><identifier>EISSN: 2381-8549</identifier><identifier>EISBN: 9781424479948</identifier><identifier>EISBN: 1424479940</identifier><identifier>EISBN: 1424479932</identifier><identifier>EISBN: 9781424479931</identifier><identifier>DOI: 10.1109/ICIP.2010.5653435</identifier><language>eng</language><publisher>IEEE</publisher><subject>Automatic voltage control ; Bandwidth ; Cache memories ; Decoding ; H.264/AVC ; Motion compensation ; Pixel ; SDRAM ; SDRAM bandwidth compression ; Tiles ; Video coding processing</subject><ispartof>2010 IEEE International Conference on Image Processing, 2010, p.3745-3748</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/5653435$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,2058,27925,54920</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/5653435$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Joon-Ho Song</creatorcontrib><creatorcontrib>Doo Hyun Kim</creatorcontrib><creatorcontrib>Do-Hyung Kim</creatorcontrib><creatorcontrib>Shi Hwa Lee</creatorcontrib><title>High-performance memory interface architecture for high-definition video coding application</title><title>2010 IEEE International Conference on Image Processing</title><addtitle>ICIP</addtitle><description>This paper proposes new memory interface architecture to overcome huge SDRAM bandwidth requirements of video decoders. To improve the memory efficiency, tile based memory access method and pixel cache are adopted. In addition to large data transfer cycle, extra overhead cycles takes great part of the heavy memory bandwidth requirement. The overhead cycles are caused by two reasons- the one is due to row-activation overhead of SDRAM and the other is due to redundant data transfer for motion compensation. The row-activation overhead is incurred frequently because of the block by block memory access pattern of standard video decoding algorithms. By using the tile based memory access method, the row-activation overhead can be reduced significantly. The redundant data transfer overhead is due to the 6-tap FIR filtering for the motion compensation while H.264/AVC decoding. We reduced the redundant memory transfer overhead by adopting the reference pixel cache, which is tightly coupled with multi-frame reference picture and the memory structure of the tile based memory access method. Experimental results show that the proposed method can decode 1920 × 1080, 30 frame per second H.264/AVC main profile bitstream in real time with only about 90MHz SDRAM clock. Compared to the conventional linear memory access system, the proposed method reduces about 6 times of SDRAM bandwidth. We also expect the proposed method will reduce power consumption because the SDRAM bandwidth requirement and the number of row-activation overhead are significantly reduced.</description><subject>Automatic voltage control</subject><subject>Bandwidth</subject><subject>Cache memories</subject><subject>Decoding</subject><subject>H.264/AVC</subject><subject>Motion compensation</subject><subject>Pixel</subject><subject>SDRAM</subject><subject>SDRAM bandwidth compression</subject><subject>Tiles</subject><subject>Video coding processing</subject><issn>1522-4880</issn><issn>2381-8549</issn><isbn>9781424479924</isbn><isbn>1424479924</isbn><isbn>9781424479948</isbn><isbn>1424479940</isbn><isbn>1424479932</isbn><isbn>9781424479931</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2010</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><sourceid>RIE</sourceid><recordid>eNpVULtOwzAUNS-JUvoBiMU_kOJrXyfOiCqgkSrBABND5cdNa9Q85Aak_j2p6MJ0dJ7DYewOxBxAlA_VonqbSzFSnWuFSp-xWVkYQIlYlCWaczaRykBmNJYX_zyJl2wCWsoMjRHX7Ga__xJi3FIwYZ_LuNlmPaW6S41tPfGGmi4deGyHUbSjYJPfxoH88J2IjzG-PVYC1bGNQ-xa_hMDddx3IbYbbvt-F709Grfsqra7Pc1OOGUfz0_vi2W2en2pFo-rLEqEIXMqkA7WBe2lza0GpwonhQpCloHAmVzlNZoigIOgC_QenQDvRIEmOKnUlN3_7UYiWvcpNjYd1qef1C8VdFl4</recordid><startdate>201009</startdate><enddate>201009</enddate><creator>Joon-Ho Song</creator><creator>Doo Hyun Kim</creator><creator>Do-Hyung Kim</creator><creator>Shi Hwa Lee</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope></search><sort><creationdate>201009</creationdate><title>High-performance memory interface architecture for high-definition video coding application</title><author>Joon-Ho Song ; Doo Hyun Kim ; Do-Hyung Kim ; Shi Hwa Lee</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i241t-b3de5dabd5c2a6a51b37b203d029de1b8636f487d1b1d574cc4b01cb0748db233</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2010</creationdate><topic>Automatic voltage control</topic><topic>Bandwidth</topic><topic>Cache memories</topic><topic>Decoding</topic><topic>H.264/AVC</topic><topic>Motion compensation</topic><topic>Pixel</topic><topic>SDRAM</topic><topic>SDRAM bandwidth compression</topic><topic>Tiles</topic><topic>Video coding processing</topic><toplevel>online_resources</toplevel><creatorcontrib>Joon-Ho Song</creatorcontrib><creatorcontrib>Doo Hyun Kim</creatorcontrib><creatorcontrib>Do-Hyung Kim</creatorcontrib><creatorcontrib>Shi Hwa Lee</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Joon-Ho Song</au><au>Doo Hyun Kim</au><au>Do-Hyung Kim</au><au>Shi Hwa Lee</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>High-performance memory interface architecture for high-definition video coding application</atitle><btitle>2010 IEEE International Conference on Image Processing</btitle><stitle>ICIP</stitle><date>2010-09</date><risdate>2010</risdate><spage>3745</spage><epage>3748</epage><pages>3745-3748</pages><issn>1522-4880</issn><eissn>2381-8549</eissn><isbn>9781424479924</isbn><isbn>1424479924</isbn><eisbn>9781424479948</eisbn><eisbn>1424479940</eisbn><eisbn>1424479932</eisbn><eisbn>9781424479931</eisbn><abstract>This paper proposes new memory interface architecture to overcome huge SDRAM bandwidth requirements of video decoders. To improve the memory efficiency, tile based memory access method and pixel cache are adopted. In addition to large data transfer cycle, extra overhead cycles takes great part of the heavy memory bandwidth requirement. The overhead cycles are caused by two reasons- the one is due to row-activation overhead of SDRAM and the other is due to redundant data transfer for motion compensation. The row-activation overhead is incurred frequently because of the block by block memory access pattern of standard video decoding algorithms. By using the tile based memory access method, the row-activation overhead can be reduced significantly. The redundant data transfer overhead is due to the 6-tap FIR filtering for the motion compensation while H.264/AVC decoding. We reduced the redundant memory transfer overhead by adopting the reference pixel cache, which is tightly coupled with multi-frame reference picture and the memory structure of the tile based memory access method. Experimental results show that the proposed method can decode 1920 × 1080, 30 frame per second H.264/AVC main profile bitstream in real time with only about 90MHz SDRAM clock. Compared to the conventional linear memory access system, the proposed method reduces about 6 times of SDRAM bandwidth. We also expect the proposed method will reduce power consumption because the SDRAM bandwidth requirement and the number of row-activation overhead are significantly reduced.</abstract><pub>IEEE</pub><doi>10.1109/ICIP.2010.5653435</doi><tpages>4</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier ISSN: 1522-4880
ispartof 2010 IEEE International Conference on Image Processing, 2010, p.3745-3748
issn 1522-4880
2381-8549
language eng
recordid cdi_ieee_primary_5653435
source IEEE Electronic Library (IEL) Conference Proceedings
subjects Automatic voltage control
Bandwidth
Cache memories
Decoding
H.264/AVC
Motion compensation
Pixel
SDRAM
SDRAM bandwidth compression
Tiles
Video coding processing
title High-performance memory interface architecture for high-definition video coding application
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-23T17%3A31%3A28IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=High-performance%20memory%20interface%20architecture%20for%20high-definition%20video%20coding%20application&rft.btitle=2010%20IEEE%20International%20Conference%20on%20Image%20Processing&rft.au=Joon-Ho%20Song&rft.date=2010-09&rft.spage=3745&rft.epage=3748&rft.pages=3745-3748&rft.issn=1522-4880&rft.eissn=2381-8549&rft.isbn=9781424479924&rft.isbn_list=1424479924&rft_id=info:doi/10.1109/ICIP.2010.5653435&rft_dat=%3Cieee_6IE%3E5653435%3C/ieee_6IE%3E%3Curl%3E%3C/url%3E&rft.eisbn=9781424479948&rft.eisbn_list=1424479940&rft.eisbn_list=1424479932&rft.eisbn_list=9781424479931&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=5653435&rfr_iscdi=true