淡江大學機構典藏:Item 987654321/35737
English  |  正體中文  |  简体中文  |  全文笔数/总笔数 : 62805/95882 (66%)
造访人次 : 3933258      在线人数 : 466
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
搜寻范围 查询小技巧:
  • 您可在西文检索词汇前后加上"双引号",以获取较精准的检索结果
  • 若欲以作者姓名搜寻,建议至进阶搜寻限定作者字段,可获得较完整数据
  • 进阶搜寻


    jsp.display-item.identifier=請使用永久網址來引用或連結此文件: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/35737


    题名: 應用於H.264/SVC之適應性正向錯誤修正與交錯式順序機制於無線網路之研究
    其它题名: A novel adaptive FEC and interleaving architecture for H.264/SVC wireless video transmission
    作者: 黃柏昌;Huang, Po-chang
    贡献者: 淡江大學電機工程學系碩士班
    李維聰;Lee, Wei-tsong
    关键词: 正向錯誤修正機制;交錯式順序機制;H.264/AVC;適應性編碼技術;無線網路;Forward Error Correction;Interleaving;H.264/AVC;Scalable Video Coding;Wireless network
    日期: 2009
    上传时间: 2010-01-11 07:05:22 (UTC+8)
    摘要: 隨著無線網路系統的快速發展,高頻寬的無線網路系統也隨之發展,但於無線的網路環境之中,網路的品質則深受環境與天候的影響。當無線網路的傳輸通道本身受到外界的影響而造成訊號衰減或干擾時,就會造成封包的遺失,因為這些因素所造成的封包遺失稱為無線遺失。當無線遺失情形嚴重時,對於偏重即時性之相關服務會是相當大的影響,如:影像串流服務便是其中之一。H.264/AVC影像編碼技術藉由參考前後之影像畫面,以提高影像的壓縮率與降低影像串流時所消耗的頻寬量。但因為此特性,當有一畫面遺失時就可能連帶影響前後數個畫面之品質與解壓縮。
    傳統的錯誤回復機制對於即時回復錯誤的封包資訊的能力有限。在許多的研究上,都致力於解決這問題。例如,正向錯誤修正機制(FEC)與自動重傳請求協議(ARQ)。正向錯誤修正機制的原理是將冗餘的資訊加在原始的封包資料之後,並利用此冗餘的資訊來回復錯誤或遺失的原始資訊。雖然正向錯誤修正機制比起自動重傳請求協議排除了時間上的延遲,但是卻使用了較多的頻寬資源。
    因此近來,Enhanced Adaptive FEC (EAFEC)被提出,用以改善影像串流在無線網路上的傳輸效能。EAFEC基於網路流量與無線頻道的狀態,來動態決定冗餘資訊之長度,以減少不必要之冗餘資訊的傳輸,並達成更有效使用無線網路頻寬資源之目的。
    於本研究中,我們將利用正向錯誤修正技術的特性,配合適應性編碼技術的層級式影像編碼架構與交錯式順序方法來延伸增強EAFEC機制之效能。此一新的機制,對於不同重要性的影像資訊能提供不同的保護強度。例如,較重要之資料給予較強之保護強度,以保證其資料的正確性;而重要程度相對低之資料則給予相對低之保護強度。此外我們提出之新機制也改善了以往遭遇連續性封包遺失時,資料的成功修復率。在此研究中,我們的主要目標是依據不同重要性的影像資料,結合網路流量和無線頻道的狀態。分別動態調整各個不同重要性資料的資料保護強度,以能更有效率的利用無線網路之頻寬資源。
    The main challenge of wireless video transmission originates from the error-prone nature caused by the time-varying channel itself. By complex calculation of preceding and succeeding frames, the H.264 video compression standard achieves high compression ratios. Packet loss degrades its quality and even affects the decoding of dependent frames. This means that the error spreads to the neighboring frames as well which depend on the several preceding differentially coded frames.
    The unavoidable wireless video transmission errors make it hard for the traditional error recovery techniques to recover the lost video packets timely. Many studies have tried to solve this problem and their major design consideration has been given to the total throughput. Such studies include Forward Error Correction (FEC) and Automatic Retransmission ReQuest (ARQ). In FEC approach, the source node transmits the parity packets along with the original data packets. The receiver can accurately recover any lost data packets less than the parity packets. The amount of the parity packet is determined at the time of FEC encoding. Although eliminating the need for time-consuming acknowledgement and retransmission operations of ARQ, FEC consumes more bandwidth.
    Recently, an intelligent FEC mechanism, Enhanced Adaptive FEC (EAFEC), has been proposed to provide improved video delivery over wireless networks. Based on both network traffic load and wireless channel state, the redundant FEC packets are dynamically added. Instead of adding unnecessary packets into the congested network, EAFEC algorithm tunes FEC packet numbers in such a way.
    In this research, we extend the EAFEC scheme to further propose and analyze the effeciency of using different FEC strength depending on H.264/SVC video stream priorities for wireless video transmission. In this scheme, we define different threshold groups to compute the FEC strength (the number of redundant packets) for the specified H.264/SVC video stream priorities with the EAFEC scheme. To protect data from the burst loss between the Access Point (AP) and the client nodes, the interleaving technique for each H.264/SVC FEC video data is also introduced. The adaptive FEC strength to different priority video frame under current rate of wireless channel loss is the objective of this research.
    显示于类别:[電機工程學系暨研究所] 學位論文

    文件中的档案:

    档案 大小格式浏览次数
    0KbUnknown429检视/开启

    在機構典藏中所有的数据项都受到原著作权保护.

    TAIR相关文章

    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - 回馈