針對多媒體行動裝置之動態服務品質視訊串流系統

隨著多媒體行動裝置逐漸變成主流消費產品，在這類行動裝置上，隨需隨看視訊串流越來越流行。然而行動裝置是以電池作為驅動的，電池效力有限，所以在設計這類視訊串流系統時，如何妥善地利用行動裝置上的電力，並且同時滿足使用者服務品質上的需求，是設計視訊串流系統的重要考量因素。在這篇論文當中，我們試著設計一個針對多媒體行動裝置，能源考量之動態服務品質視訊串流系統。我們也另外設計了一個影片內容考量之視訊串流耗能預測器，來估計在行動裝置上視訊串流可能耗電。在所設計的動態服務品質視訊串流，我們將一段影片切成一段段等長的影片小段，並將這些影片小段以不同的比特率來壓縮，藉由所設計之視訊串流預測器，以及已知每個影片小段之服務品質，我們使用動態規劃來挑選每個影片小段適當的比特率組成一組完整的影片，這段影片可以滿足使用者所給定的服務品質需求，並且在其行動裝置上播放耗能可達最小。我們使用一個搭載英特爾微處理器之實驗開發板作為行動裝置，並且透過無線網路傳輸和一台主機作視訊串流來模擬所設計之視訊串流系統。藉由實際測量實驗開發板之耗能，所設計之視訊串流系統可以節省高達 9.08% 之耗能，並同時滿足使用者所給定的服務品質需求。

關鍵字

能源考量；視訊串流；耗能預測；動態服務品質

並列摘要

As multimedia mobile devices become mainstream consumer products, an energy-aware on-demand video streaming system for portable devices is critical. Since the battery life time on multimedia mobile devices is limited, how to utilize energy efficiently while meeting the required quality of service (QoS) is a major concern in designing a on-demand video streaming system. In this thesis, we propose an energy-aware video quality adaptation mechanism for energy-aware video streaming system. We also devise a content-aware energy predictor to predict the possible energy consumption for consuming the required service. With the aid of the proposed energy predictor, dynamic programming method is used to generate an optimal video stream composed by multiple bitstreams with different bit-rates to minimize total energy consumption on the mobile device while meeting the user-defined QoS requirements. We use an Intel XScale-based evaluation board as a mobile device in our streaming system in which a server and the evaluation platform communicate with each other through wireless connection. According to the actual energy measurement in the evaluation platform, the proposed system can achieve up to 9.08% energy reduction while guaranteeing the user-defined QoS requirements.

並列關鍵字

energy-aware ； video streaming ； energy prediction ； adaptive QoS

參考文獻

[3] K. Choi, R. Soma, and M. Pedram. Off-chip latency-driven dynamic voltage and frequency scaling for an mpeg decoding. In DAC ’04: Proceedings of the 41st annual conference on Design automation, pages 544–549, New York, NY, USA, 2004. ACM.

[4] G. Contreras and M. Martonosi. Power prediction for intel xscale processors using performance monitoring unit events. In ISLPED ’05: Proceedings of the 2005 International symposium on Low power electronics and design, pages 221–226, New York, NY, USA, 2005. ACM.

[5] V. Devadas and H. Aydin. On the interplay of dynamic voltage scaling and dynamic power management in real-time embedded applications. In EMSOFT ’08: Proceedings of the 8th ACM international conference on Embedded software, pages 99–108, New York, NY, USA, 2008. ACM.

[7] Y. Huang, A. V. Tran, and Y. Wang. A workload prediction model for decoding mpeg video and its application to workload-scalable transcoding. In MULTIMEDIA ’07: Proceedings of the 15th international conference on Multimedia, pages 952–961, New York, NY, USA, 2007. ACM.

[8] Intel Corporation. Intel PXA27x Processor Family: Developer’s Manual, October 2004. Order No. 280000-002.

國際替代計量

針對多媒體行動裝置之動態服務品質視訊串流系統

全文下載

主題瀏覽