Vol.54 No (Nov. 2013) P2P 1,a) 1,b) 1,c) 1,d) , P2P CG P2P A Method to Reduce Interruption Time of Model and Motion Sep

Size: px

Start display at page:

Download "Vol.54 No (Nov. 2013) P2P 1,a) 1,b) 1,c) 1,d) , P2P CG P2P A Method to Reduce Interruption Time of Model and Motion Sep"

ときなすみい
4 years ago
Views:

1 P2P 1,a) 1,b) 1,c) 1,d) , P2P CG P2P A Method to Reduce Interruption Time of Model and Motion Separated Contents in P2P Streaming Environments Masahiro Yokoyama 1,a) Tomoki Yoshihisa 1,b) Takahiro Hara 1,c) Shojiro Nishio 1,d) Received: January 7, 2013, Accepted: September 13, 2013 Abstract: In recent years, there has been an increasing interest in P2P (Peer-to-Peer) streaming environments and delivered contents are diversified. A typical example is a content that can be separated into model and motion data like CG animations. Although the interruption time for playing the data can be reduced by considering the time to start playing each divided data, previous researches do not focus on such contents. In this paper, we propose a method to reduce interruption time of model and motion separated contents in P2P streaming environments. The evaluation results show that our proposed method can reduce the interruption time for peers when requests for playing the data burstly arrive. Keywords: video-on-demand, streaming data, webcasting 1. P2P Peer-to- 1 Graduate School of Information Science and Technology, Osaka University, Suita, Osaka , Japan a) yokoyama.masahiro@ist.osaka-u.ac.jp b) yoshihisa@ist.osaka-u.ac.jp c) hara@ist.osaka-u.ac.jp d) nishio@ist.osaka-u.ac.jp Peer P2P P2P P2P DICOMO c 2013 Information Processing Society of Japan 2360

2 TVML TV program Making Language *1 MMD MikuMikuDance CG [5], [7], [10] P2P 3 3 P2P P2P P2P P2P *1 TVML, 2. P2P Bit- Torrent [1] *2 BitTorrent P2P BitTorrent BiToS Enhancing BitTorrent for Supporting Streaming Applications [11] BIS-S BiToS + Immediacy and Scarcity based Streaming [8] BitTorrent BIS-S 1 p I S c BIS-S P2P FLoD Flowing Level-of-Details [5] 3D AOI Area of Interest 3D 1 BIS-S Fig. 1 A priority set and a non-priority set for the BIS-S method. *2 BitTorrent, c 2013 Information Processing Society of Japan 2361

3 P2P 2 P2P P2P P2P P2P P2P P2P CG 2 P2P Fig. 2 Our assumed P2P streaming environment. 1 CG TVML 4. RP-ET Randomize + Prefetch considering Extra Time P2P 1 RP-ET E t s c 2013 Information Processing Society of Japan 2362

4 c 1 d E = s + dc t (1) s = t c =0 E =0 T 1 T 2 T 1 <T 2 E T 1 T 1 <E T 2 E w w w = E T 1 U (2) T 2 T 1 U T 1 =10 T 2 =30 E =20 U =4 w =2 2 T 2 <E 4.2 RP-ET 2 T 1 2 T 2 T 1 2 T 1 = 2D R,T 2 = nd (n =3, 4, ) (3) R D R 1 n 1 1 p B p N p M p p R R = B p N p + M p Q q Q 4.2 Bq N q+m q+1 max q Q B q N q + M q +1 (4) (5) N q M q 1 RP-ET N q M q Q RP-ET c 2013 Information Processing Society of Japan 2363

5 1 Table 1 Experimental parameters. 3 RP-ET Fig. 3 An example of the RP-ET method. ID A5 ID E 20 T 1 T (2) w 2 2 A4 A5 A5 A T 1 <E T 2 E T 1 A4 T 2 <E 2 M9 M16 5. RP-ET LAN 6 Mbps 1 Mbps [ ] 1,800 [ ] 180 [Mbps] 2.5 [%] 75 D [Kbyte] 1,024 [ ] 20 [ ] 1 B [Mbps] 4 8 M [ ] 2 N [ ] 2 4 Mbps 8 Mbps 1 10 BitTorrent 1MByte MPEG2 2.5 Mbps % T 2 n , [3] 1, % P2P GPS [12] GPS P2P [6], [9] c 2013 Information Processing Society of Japan 2364

6 5.2 BIS-S RP-ET [8] P2P BIS-S BIS-S 5.3 T 2 RP-ET T 2 n n 100 ID ID 2,751 4 n T 2 1 T 2 75% 80% 60% 1 5 n n RP-ET n T 2 Fig. 4 The influence of the model-data ratio and T 2. 5 T 2 Fig. 5 The influence of the scene durations and T 2. c 2013 Information Processing Society of Japan 2365

7 6 1.5 Fig. 6 Interruption time in transition phase (The arrival interval is 1.5 sec.). 2 BIS-S RP-ET P2P ID Fig. 7 Interruption time in transition phase (The arrival interval is 2 sec.). 2 [ ] Table 2 The convergence time and the arrival interval (sec.) RP-ET BIS-S ID ID2751 P2P RP-ET RP-ET BIS-S ID BIS-S P2P 5.6 c 2013 Information Processing Society of Japan 2366

8 8 Fig. 8 2 Interruption time in transition phase (Scene duration is 2 min.) % Fig. 10 Interruption time in transition phase (Model-data ratio is 60%). 9 3 Fig. 9 Interruption time in transition phase (Scene duration is 3 min.). 3 [ ] Table 3 The convergence time and the scene durations (sec.). 2 3 RP-ET , , BIS-S % 60% 80% RP-ET BIS-S 11 80% Fig. 11 Interruption time in transition phase (Model-data ratio is 80%). 4 [ ] Table 4 The convergence time and model-data ratio (sec.). 60% 80% RP-ET , BIS-S P2P c 2013 Information Processing Society of Japan 2367

9 Table 5 5 [ ] The convergence time and model-data ratio (sec.). RP-ET , , , BIS-S Fig. 12 Interruption time in transition phase (Model-data ratio diminishes linealy). RP-ET Fig. 13 Interruption time in transition phase (Model-data ratio is constant) % 20% 50% RP-ET BIS-S 6.1 RP-ET 50 RP-ET RP-ET P2P P2P RP-ET 6.2 T 2 T 2 T 2 RP-ET T 2 c 2013 Information Processing Society of Japan 2368

6.3 [2] [4] 7. P2P RP-ET RP-ET A 23680007 SCOPE 23650050 [1] Cohen, B.: Incentives build robustness in BitTorrent, Proc. P2PECON, pp.68 72 (2003). [2] Chim, J., Lau, R., Leong, H.V. and Si, A.

10 6.3 [2] [4] 7. P2P RP-ET RP-ET A SCOPE [1] Cohen, B.: Incentives build robustness in BitTorrent, Proc. P2PECON, pp (2003). [2] Chim, J., Lau, R., Leong, H.V. and Si, A.: CyberWalk: A Web-based Distributed Virtual Walkthrough Environment, IEEE Trans. Multimedia, Vol.5, No.4, pp (2003). [3] D Acunto, L., Vinko, T. and Pouwelse, J.: Do BitTorrent-like VoD systems scale under flash-crowds?, Proc. IEEE Int. Conf. on P2P, pp.1 4 (2010). [4] Doran, A., Mondet, S., Grigoras, R., Morin, G., Ooi, W.T. and Boudon, F.: A Demonstration of Mobi- Tree: Progressive 3D Tree Models Streaming on Mobile Clients, Proc. ACM Multimedia, pp (2009). [5] Hu, S.Y., Huang, T.H., Chang, S.C., Sung, W.L., Jiang, J.R. and Chen, B.Y.: Flod: A framework for peer-topeer 3D streaming, Proc. IEEE INFOCOM, pp (2008). [6] Kangasharju, J., Schmidt, U., Bradler, D. and Schröder-Bernhardi, J.: ChunkSim: Simulating peer-topeer content distribution, Proc. Spring Simulaiton Multiconference, pp (2007). [7] Vol.49, No.2, pp (2008). [8] P2P Vol.52, No.11, pp (2011). [9] Shah, P., Rasheed, J. and Paris, J.F.: Performance study of unstructured P2P overlay streaming systems, Proc. Int. Conf. on Computer Communications and Networks, pp.1 6 (2008). [10] Tang, Z., Rong, G., Guo, X. and Prabhakaran, B.: Streaming 3D Shape Deformations in Collaborative Virtual Environment, Conf. IEEE VR, pp (2010). [11] Vlavianos, A., Iliofotou, M. and Faloutsos, M.: BiToS: Enhancing BitTorrent for supporting streaming applications, Proc. IEEE INFOCOM, pp.1 6 (2006). [12] Yang, W. and Ghazaleh, N.A.: GPS: A general peer-topeer simulator and its use for modeling BitTorrent, Proc. IEEE Int. Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, pp (2005) P2P c 2013 Information Processing Society of Japan 2369

11 IEEE IEEE ACM IEEE c 2013 Information Processing Society of Japan 2370

& Vol.5 No (Oct. 2015) TV 1,2,a) , Augmented TV TV AR Augmented Reality 3DCG TV Estimation of TV Screen Position and Ro

& Vol.5 No (Oct. 2015) TV 1,2,a) , Augmented TV TV AR Augmented Reality 3DCG TV Estimation of TV Screen Position and Ro TV 1,2,a) 1 2 2015 1 26, 2015 5 21 Augmented TV TV AR Augmented Reality 3DCG TV Estimation of TV Screen Position and Rotation Using Mobile Device Hiroyuki Kawakita 1,2,a) Toshio Nakagawa 1 Makoto Sato