VBR视频流量的小波包分解及其长时预测

doi:1000-436X(2008)06-0034-09

摘要/Abstract

摘要：

长时预测是VBR 视频流量预测领域中的难点问题。针对其时变、非线性以及长相关性等特点，提出一种多尺度分解的 VBR 视频业务的特征提取方法。选择具有任意多分辨分解特性的小波包，对其进行空间划分并求解适合视频信号特征提取的最优分解基。基于最优基对视频信号进行快速多尺度分解，得到了各级节点的小波系数矩阵，建立了基于最小二乘支持向量机与最小均方的小波系数预测方法。最后，根据预测小波系数，进一步提出了基于小波系数逆变换的视频流量长时预测方法。仿真结果验证了此算法的有效性。

关键词: VBR业务流量, 多尺度分解, 小波包, 长时预测, 最小二乘支持向量机, 最小均方预测

Abstract:

Long-term prediction is one of the most difficult problems in the area of VBR video traffic prediction.As to the time variation,non-linearity and long range dependence in VBR video traffic trace,a novel method of feature based on multi-scale decomposition was proposed.On the analysis of the time-frequency distribution characteristics of the video trace,the wavelet packets which have the trait of arbitrary distinction and decomposition are selected.After space partition of wavelet packets,the best wavelet packet basis for feature extraction is picked out.Based on the best basis,it can do fast arbitrary multi-scale WPT (wavelet packet transform),and obtain each higher dimension wavelet coefficients matrix.And then wavelet coefficients prediction is proposed based on LS-SVM and LMS algorithms.The long-term pre-diction of VBR video traffic is obtained through reverse wavelet transforms on the predicted wavelet coefficients.Nu-merical and simulation results are provided to validate the claims.

Key words: VBR video traffic, multi-scale decomposition, wavelet packet, long-time prediction, LS-SVM, LMS prediction

陈剑,闻英友,赵大哲,刘积仁. VBR视频流量的小波包分解及其长时预测[J]. 通信学报, 2008, 29(6): 34-43.

Jian CHEN,Ying-you WEN,Da-zhe ZHAO,Ji-ren LIU. Long-term prediction for VBR video traffic based on wavelet packet decomposition[J]. Journal on Communications, 2008, 29(6): 34-43.

图/表 6

图1

图2

表1

图3

表2

图4

参考文献 33

[1]	LAZAR A A , PACIFICI G . Control of resources in broadband net-works with quality of service guarantees[J]. IEEE Communications, 1991,29(10): 66-73.
[2]	CHONG S , LI S , GHOSH J . Predictive dynamic bandwidth allocation for efficient transport of real-time VBR video over ATM[J]. IEEE Journal on Selected Areas in Communication, 1995,13(1): 12-23.
[3]	ALARCON-AQUINO V , BARRIA J A . Multiresolution FIR neu-ral-network-based learning algorithm applied to network traffic prediction[J]. IEEE Transactions on Systems,Man,and Cybernetics, 2006,36(2): 208-220.
[4]	LELAND W E , WILLINGER W , TAQQU M S ,et al. On the self-similar nature of Ethernet traffic[A]. Proc of the ACM SIG-COMM[C]. San Francisco,California, 1993. 183-193.
[5]	LELAND E , TAQQU M S , WILLINGER W ,et al. On the self similar nature of Ethernet traffic (extended version)[J]. IEEE/ACM Transac-tions on Networking, 1994,2(1): 1-15.
[6]	BERAN J , SHERMAN R , TAQQU M S ,et al. Long-range depend-ence in variable-bit-rate video traffic[J]. IEEE Transactions on Communication, 1995,43(234): 1566-1579.
[7]	PAXSON V , FLOYD S . Wide area traffic:the failure of Poisson modeling[J]. IEEE/ACM Transactions on Networking, 1995,3(3): 226-244.
[8]	RATHGEB E P . Modeling and performance comparison of policing mechanism for ATM network[J]. IEEE Journal on Selected Areas in Communication, 1991,9(3): 325-334.
[9]	CROVELLA M , BESTAVROS A . Self similarity in World Wide Web traffic:evidence and possible causes[J]. IEEE/ACM Transactions on Networking, 1997,5(6): 835-846.
[10]	TSYBKOV B , GEORGANAS N D . On self-similar traffic in ATM queues:definitions,overflow probability bound,and cell delay distri-bution[J]. IEEE/ACM Transactions on Networking, 1997,5(3): 397-409.
[11]	BERAN J , SHERMAN R , TAQQU M S ,et al. Long-range depend-ence in variable-bit-rate video traffic[J]. IEEE Transactions on Com-munications, 1995,4(3): 1565-1579.
[12]	ATIYA A F , ALY M A , PARLOS A G . Sparse basis selection:new results and application to adaptive prediction of video source traf-fic[J]. ·IEEE Transactions on Neural Network, 2005,16(5): 1136-1146.
[13]	LIANG Y , PAGE E W . Multiresolution learning paradigm and signal prediction[J]. IEEE Transactions on Signal Processing, 1997,45(11): 2858-2864.
[14]	LIANG Y . Multi-Resolution Learning Paradigm and High-Speed Network Traffic Prediction[D]. Department of Computer Science,Clemson University,USA, 1997.
[15]	LIANG Y , PAGE E W . VBR video traffic prediction using neural networks with multi-resolution learning[A]. Advances in Neural Net-works and Applications[C]. Athens,Greece, 2001. 98-103.
[16]	GARRETT M W , WILLINGER W . Analysis,modeling and genera-tion of self-similar VBR video traffic[A]. Proc of the ACM SIG-COMM[C]. London, 1994. 269-280.
[17]	TSE D , GALLAGER R , TSITSIKLIS J . Statistical multiplexing of multiple time scale Markov streams[J]. IEEE Journal on Selected Areas in Communication, 1995,13(6): 1028-1038.
[18]	SANG A , LI S . A predictability analysis of network traffic[A]. Proc of the 19thAnnual Joint Conference of the IEEE Computer and Communications Societies (INFOCOM)[C]. Tel-Aviv,Israel, 2000. 342-351.
[19]	MALLAT S . A theory for multi-resolution signal decomposition:the wavelet representation[J]. IEEE Transactions on Pattern Analysis Machine Intelligence, 1989,11(7): 257-263.
[20]	COHEN A , KOVACEVIC J . Wavelets:the mathematical background[J]. Proceedings of the IEEE, 1996,84(4): 514-522.
[21]	TEWFIK A H , KIM M . Correlation structure of the discrete wavelet coefficients of fractional Brownian motion[J]. IEEE Transactions on Information Theory, 1992,38(2): 904-909.
[22]	WILLINGER W , TAQQU M S , SHERMAN R . Self-similarity through high-variability:statistical analysis of Ethernet LAN traffic at source level[J]. IEEE/ACM Transactions on Networking, 1997,5(1): 71-86.
[23]	PATRICK F L . Wavelet analysis and synthesis of fractional Brownian mo-tion[J]. IEEE Transactions on Information Theory, 1992,38(2): 910-917.
[24]	SARVOTHAM S , RIEDI R , BARANIUK R . Connection level analy-sis and modeling of network traffic[A]. Proc IEEE/ACM Network Measurement Workshop[C]. San Francisco, 2001. 99-103.
[25]	MALLAT S . A Wavelet Tour of Signal Processing (Second Edi-tion)[M]. Washington: Academic Press, 1999.
[26]	王首勇, 朱光喜, 唐远炎 . 应用最优小波包变换的特征提取方法[J]. 电子学报, 2003,31(7): 1035-1038. WANG S Y , ZHU G X , TANG Y Y . Feature extraction using best wavelet packet transform[J]. Chinese Journal of Electronics, 2003,31(7): 1035-1038.
[27]	刘毅, 张彩明, 赵玉华 et al. 应用最优小波包变换的特征提取方法[J]. 计算机学报, 2006,29(5): 769-777. LIU Y , ZHANG C M , PENG Y H ,et al. The feature extraction and classification of lung sounds based on wavelet packet multi-scale analysis[J]. Chinese Journal of Computer, 2006,29(5): 769-777.
[28]	CHIRUVOLU G , SANKAR R , RANGANATHAN N . Adaptive VBR traffic management for higher utilization of ATM networks[J]. ACM SIGCOMM Computer Communications Review, 1998,28(3): 27-40.
[29]	ADAS A . Using adaptive linear prediction to support real-time VBR video under RCBR network service model[J]. IEEE Transactions on Networking, 1998,6(5): 635-644.
[30]	IERA A , MOLINARO A , MARANO S . Call admission control and resource management issues for real-time VBR traffic in ATM-satellite networks[J]. IEEE Journal on Selected Areas in Com-munications, 2000,18(11): 2393-2403.
[31]	JOHAN A K S , TONY V G , JOS D B . Least squares support vector machines[J]. Machine Learning, 2002,(7): 1-25.
[32]	WEIGEND A S , GERSHENFELD N A . Time Series Prediction:Forecasting the Future and Understanding the Past[M]. Redwood City: Addition-Wesley Publishing, 1994.
[33]	苏晓星, 常胜江, 熊涛等. 用神经网络实现 VBR 视频通信量的在线预测[J]. 电子学报, 2005,33(7): 1163-1167. SU X X , CHANG S J , XIONG T ,et al. Online VBR video traffic pre-diction using neural network[J]. Chinese Journal of Electronics, 2005,33(7): 1163-1167.

小波包层次	小波系数长度	需要预测小波系数长度
AAA3	128	28
DAA3	128	28
DA2	256	56
D1	512	112

帧类型	50帧		100帧		120帧		150帧		200帧
帧类型	LS-SVM	LMS	LS-SVM	LMS	LS-SVM	LMS	LS-SVM	LMS	LS-SVM	LMS
StarWars	0.016	0.078	0.063	0.157	0.112	0.195	0.156	0.287	0.253	0.413
News	0.055	0.126	0.175	0.297	0.247	0.365	0.398	0.523	0.523	0.745