應用於可調式影像編碼之網路頻寬適應性多媒體串流系統設計

Design of Bandwidth Adaptive Streaming System for Scalable Video Coding

蕭詣懋(Yi-Mao Hsiao)

指導教授 ： 朱元三

國立中正大學/工學院/電機工程學系/博士(2012年)

摘要

在網際網路上傳送多媒體串流時所遭遇到的問題：伺服器端要如何同時支援多種不同規格的影像格式，及面對不同的用戶端處理器能力和網路頻寬。此外，由於網路流量日益繁重，骨幹網路必須能夠更快速處理路由查找問題，以提升網際網路上多媒體串流的傳送效率。可調式視訊編碼(H.264/SVC)視訊串流機制的提出，其可同時壓縮多種不同規格的影像格式。但當網路環境變化時，如頻寬壅塞時，縱使有可調式視訊串流機制，亦無法發揮效能，造成視訊品質(PSNR)下降。在本論文中，我們提出了應用於可調式視訊串流之頻寬適應策略。此策略在網路發生壅塞時進行了影像位元率的調整，其中包含了Macro與Micro control。在Macro control 中，伺服器端對於異質性的使用者和網路環境分配給最適當H.264/SVC 串流影像。在Micro control 中，伺服器端對於影像位元率和網路頻寬發生變化時，採取主動位元率的控制。我們針對提出的策略實作出可調式視訊串流系統，由實作結果得知，以Foreman和Stefan影像可提升了PSNR值 4.01%到10.89%。此外，我們提出一個路由查找系統來做使用者連線管理。系統裡面涵蓋了一個路由查表ASIC和外部晶片的路由查找表。外部晶片的 IPv6路由查找表是兩層的階層式記憶體架構，根據6Net路由器路由表的prefix 長度分佈所設計。第一層外部晶片路由表可以容納routing entries的91.89%，第二層則剩下的8.11%。ASIC裡面包含了一個Function unit和一個Binary CAM。Binary CAM被當成快取記憶體來使用，其可容納1024 個entries，並且保證80%的hit ratio。Binary CAM中的取代演算法使用FIFO。此路由查表系統可達260Mlps的速度換算為100Gbps的系統效能，其滿足目前網路上的需求，未來更大的頻寬亦可使用。系統中用了 20.04 KB的TCAM ，10.24KB BCAM以及 29.29MB RAM，並可容納3.6M的routing entries。我們提出的多媒體串流系統設計，可應用於網際網路上傳送即時影像的需求。針對使用者處於不論是有線或是無線的網路中，解決了可調式視訊串流在網際網路上傳送時所遇到的頻寬變化和網路壅塞的問題。此系統亦可應用於骨幹網路，解決路由查找問題以及網路流量的擁塞。

關鍵字

多媒體串流系統 ； 路由查找

並列摘要

Designing a real-time video-streaming system involves the challenges, which are network congestion and bandwidth fluctuations, of lowering the video quality. Because of the increasing heavy network traffic, the backbone network has to handle routing lookup more quickly so that the transmission efficiency of multimedia streaming on the Internet can be improved. We design and implement a real-time video-streaming system for scalable video coding (SVC) and propose an adaptive control scheme, containing macro- and micro-control, to resolve the bandwidth fluctuation and network congestion. Under macro-control, the server allocates appropriated video-streaming of SVC for heterogeneous clients based on terminal capacity and network conditions. Under micro-control, the server changes the video frame rate based on video bit rates and network bandwidth. The streaming system with the adaptive control scheme is implemented to improve the peak signal noise ratio (PSNR) by 4.01 % to 10.89 % of the Foreman and Stefan videos. A routing lookup system is presented to manage connection, and it is composed of routing lookup ASIC and off-chip memory set. The off-chip memory set is two-level hierarchical memory architecture. 91.89% routing entries of the routing table can be searched in one memory access, and the worst case about 10% in this system is two memory accesses .The ASIC includes a function unit and a Binary Content Addressable Memory (BCAM) . The BCAM is used as cache memory with FIFO replacement algorithm .There are 1024 cache entries in the BCAM with 80% hit ratio. The routing lookup system approaches 260 Mega lookups per second (Mlps), which is sufficient for 100 Gbps networks. The memory density is good, with each routing entry requiring only 64 bits. Moreover, the routing table only needs 10.24KB on-chip BCAM, 20.04KB off-chip TCAM and 29.29MB DRAM for 3.6M routing entries in the proposed system. The proposed system is designed for real-time streaming transmission over Internet to resolve the bandwidth fluctuation and network congestion. The system is also designed for backbone network to resolve routing lookup and heavy network traffic.

並列關鍵字

Multimedia streaming system ； Routing Lookup

參考文獻

[6] M. Mrak, M. Grgic and S. Grgic, “Scalable Video Coding in Network Applications,” in Proc. IEEE VIPromCom, 2002.

[7] C. H. Foh, Y. Zhang, Z. Ni and J. Cai, “Scalable Video Transmission over the IEEE 802.11e Networks Using Cross-Layer Rate Control,” in Proc. IEEE ICC, 2007.

[8] W.-T. Chen, T.-C. Lin, Y.-C. Chang and J.-C. Chen, “Dynamic Packet Selection for H.264 Video Streaming over IEEE 802.11e WLANs,” in Proc. IEEE WCNC, 2008.

[9] Y. Wang, L.-P. Chau and K.-H. Yap, “Spatial Resolution Decision in Scalable Bitstream Extraction for Network and Receiver Aware Adaptation,” in Proc. IEEE International Conference on Multimedia and Expo, June, 2008.

[10] D. T. Nguyen and J. Ostermann, “Congestion Control for Scalable Video Streaming Using the Scalability Extension of H.264/AVC,” IEEE Journal of Selected Topics in Signal Processing, vol. 1, no. 2, pp: 246 – 253, 2007.