In this dissertation, the traffic control functions involving the connection admission con-trol and the traffic policing for multimedia high-speed networks are studied by employing the neural/fuzzy intelligent techniques or a sophisticated computation algorithm, including a neural fuzzy connection admission controller, a power-spectrum-based neural-net connection admission controller, a fuzzy increment controller, a neural fuzzy increment controller and a enhanced traffic marker. Both ATM and IP networks which can be utilized to construct the multimedia high-speed networks are considered in this dissertation. The CAC schemes which make the admission control decisions according to the time-domain and frequency-domain traffic parameters are both discussed where the intelligent techniques are chosen to implement the CAC controllers. Also, the enhanced algorithms which implement the traffic policing function by incorporating the intelligent techniques and a elaborate compu-tation procedure into existing algorithms for ATM and IP networks respectively are both well explored.
In Chapter 3, a neural fuzzy connection admission control (NFCAC) scheme which based on the time-domain traffic parameters and provides QoS guarantees for ATM networks is proposed. The NFCAC scheme combines the linguistic control capability of a fuzzy logic controller and the learning ability of a neural network. This type of integrated neural fuzzy system can automatically construct a rule structure by learning from training examples
and can self-calibrate parameters of membership functions. It not only provides a robust framework to mimic experts’ knowledge embodied in existing traffic control techniques but also constructs intelligent computational algorithms for traffic control. It can be easily trained and enhances system utilization. Simulation results show that the proposed NFCAC scheme provides system utilization about 32% and 11% higher than the EBCAC and FLCAC schemes proposed in [10] and [18], respectively, and the NFCAC scheme requires only a fraction of the 103 order and the 101 order of training cycles, consumed by the NNCAC scheme proposed in [23] and RBFCAC scheme, respectively. An NFCAC scheme such as the one introduced here may be the answer to the problem of designing a coherent call admission controller for ATM systems.
In Chapter 4, we propose a power-spectrum-based neural-net connection admission con-trol (PNCAC) scheme for ATM networks. The PNCAC method adopts the converted power-spectrum parameters of traffic source to represent its traffic characteristics and uses neural network to implement the connection admission control. The frequency-domain power-spectrum parameters of traffic source possess additive property and can capture the cor-relation and burstiness behavior more than the time-domain parameters such as peak rate, mean rate, and peak rate duration. The neural network has the learning/adapting capabili-ties so that the boundary of the decision hyperplane for the connection admission control can be adjusted optimally and dynamically. Simulation results show that the proposed PNCAC enhances significantly the system utilization while fulfilling QoS requirements. Not only is it superior to the conventional equivalent capacity CAC scheme (ECCAC), it also obtains more flexibility and robustness than Hiramatsu’s NNCAC.
However, the practical traffic characteristics of multimedia services in broadband net-works may change very fast and abruptly with large volume. Also, several researches demon-strate that the multimedia traffic possesses self-similar or chaotic property, and present a
long-range dependence (LRD). The conventional traffic control algorithms based on current system performance measures may not perform well because of the fast varying dynamic traffic; the control decision would be obsolete and inappropriate due to delayed react to such a fast dynamic traffic. It is necessary to capture the next-step system performance, that is, the predicted information about the system status due to traffic change should be provided. Accordingly, a predictive intelligent traffic controller for broadband multimedia systems could be proposed as the future work for the CAC. It considers predicted system performance measures, instead of present ones, to well capture the oncoming effects in the future, besides also adopting the neural fuzzy network for the CAC decision making as well as the fuzzy logic controller for both of the equivalent capacity estimation of the new call and congestion estimation of the system. A pipelined recurrent neural network with extended recursive least square learning algorithm (PRNN/ERLS) [61], [62], which can efficiently re-duce the prediction error for the statistical fluctuations of the system, could be employed to implement the predictors to well attain the advance information of the system. It is expected that the predictive intelligent traffic controller with predicted system measured statistics would achieves better performances than that of the conventional CAC schemes without prediction.
In Chapter 5, we employ two intelligent techniques, the fuzzy logic systems and the neural fuzzy networks, to design two intelligent leaky bucket algorithms, respectively, for sustainable-cell-rate usage parameter control of multimedia transmission in ATM networks.
The first algorithm we proposed is the fuzzy leaky bucket algorithm, which as the name im-plies, employs a fuzzy increment controller (FIC) in conjunction with the conventional leaky bucket algorithm. The FIC monitors the long-term mean rate and the short-term mean rate of a connection and uses the fuzzification, inference rules and defuzzification to process them in order to derive the optimal increment value. The other intelligent leaky bucket
algorithm we proposed is the neural fuzzy leaky bucket algorithm, which utilizes a neural fuzzy increment controller (NFIC) to dynamically adjust the increment value. The NFIC is basically an FIC except that it further employs a neural network to optimize its fuzzy logic system through the reinforcement learning. Simulation results show that, regardless of the traffic sources chosen, both intelligent leaky bucket algorithms achieve better performances in terms of selectivity, responsiveness and mean queueing delay as compared to the conven-tional leaky bucket algorithm by responding about 160% faster when taking control actions against a non-conforming connection, while reducing as much as 50% of the queueing delay experienced by a conforming connection. The performance gain of the intelligent algorithms is a result of employing fuzzy logic and neural fuzzy controllers where the measured system statistics, the long-term and short-term mean rates, are introduced as the feedback infor-mation and served as the inputs of the intelligent controllers to form a robust and adaptive close-loop control system. Accordingly, both intelligent algorithms can adapt to the time-varying and non-stationary traffic, and thus enhance their performances. In addition, the simulation results also show that the neural fuzzy leaky bucket algorithm outperforms the fuzzy one by achieving better performance in all aspects especially the responsiveness.
In Chapter 6, we proposed an enhanced traffic marker (ETM) for the TRTCM-based traffic conditioner to perform the traffic policing function in DiffServ IP networks. The primary feature of the proposed ETM is that it can fairly allocate the color notations among connections within an aggregate one. It also enhances the throughput of each conforming level for the aggregate connection to achieve as high rate as possible by not only restore the conforming levels of the previously demoted packets, but also aggressively promote the packets to higher conforming levels if the network resource condition is available, so that the end-to-end QoS of the applications would be substantially improved while the traffic contract is still be respected. The operations of the ETM scheme as well as the computations of the
promotion/demotion probabilities are carefully defined. The performances of the proposed ETM scheme were verified via simulations and the simulation results were compared with the conventional TRTCM scheme. Simulation results show the ETM scheme outperform the TRTCM scheme in both aspects of marking fairness and traffic throughput of each conforming level under congested and under-loaded networks.
Bibliography
[1] C. T. Lin and C. S. G. Lee, Neural Fuzzy Systems: a neuro-fuzzy synergism to intelligent systems, Prentice–Hall, Singapore, 1996.
[2] J.-S. R. Jang, C. T. Sun and E. Mizutani, Neuro-Fuzzy and Soft Computing: a compu-tational approach to learning and machine intelligence, Prentice–Hall, 1997.
[3] H.-J. Zimmermann, Fuzzy set theory and its applications, 2nd revised edition, Kluwer Academic Publishers, pp. 11–17, 1991.
[4] C. C. Lee, “Fuzzy logic in control systems: Fuzzy logic controller – Part I,” IEEE Tran.
Systems, Man And Cybernetics, Vol. 20, No. 2, pp. 419–435, Mar./Apr. 1990.
[5] N. E. Cotter, “The Stone-Weierstrass theorem and its application to neural networks,”
IEEE Trans. Neural Networks, Vol. 1, Iss. 4, pp. 290–295, Dec. 1990.
[6] D. E. Rumelhart, G. E. Hinton, R. J. Williams, “Learning internal representation by error propagation,” Parallel Distributed Processing: Explorations in the Microstructure of Cognition, MA: MIT Press, Cambridge, Vol. 1, Ch. 8, 1986.
[7] J. Moody and C. Darken, “Fast learning in networks of locally-tuned processing units,”
Neural Computa., Vol. 1, No. 2, pp. 281-294, 1989.
[8] R. Gu`erin, H. Ahmadi and M. Naghshineh, “Equivalent capacity and its application to bandwidth allocation in high–speed networks,” IEEE J. Select. Areas Commun., Vol.
9, No. 7, pp. 968–981, Sep. 1991.
[9] C. S. Chang and J. A. Thomas, “Effective bandwidth in high-speed digital networks,”
IEEE J. Select. Areas Commun., Vol. 13, No. 6, pp. 1091–1100, Aug. 1995.
[10] G. Kesidis, J. Walrand and C. S. Chang, “Effective bandwidths for multiclass Markov fluids and other ATM sources,” IEEE/ACM Tran. Networking, Vol. 1, No. 4, pp.424–
428, Aug. 1993.
[11] A. I. Elwalid and D. Mitra, “Effective bandwidth of general Markovian traffic sources and admission control of high speed networks,” IEEE/ACM Trans. Networking, Vol. 1, No. 3, pp. 329–343, June 1993.
[12] A. I. Elwalid, D. Heyman, T. V. Lakshman, D. Mitra and A. Weiss, “Fundamental bounds and approximations for ATM multiplexers with applications to video telecon-ferencing,” IEEE J. Select. Areas Commun., Vol. 13, No. 6, pp. 1004–1016, Aug. 1995.
[13] H. Saito, “Call admission control in an ATM network using upper bound of cell loss probability,” IEEE Trans. Commun., Vol. 40, No. 9, pp. 1512–1521, Sep. 1992.
[14] M. Murata, Y. Oie, T. Suda and H. Miyahara, “Analysis of a discrete-time single-server queue with bursty inputs for traffic control in ATM networks,” IEEE J. Select. Areas Commun., Vol. 8, No. 3, pp. 447–458, April 1990.
[15] S. Jamin, P. B. Danzig, S. J. Shenker and L. Zhang, “A framework for robust measurement-based admission control,” IEEE/ACM Trans. Networking, Vol. 5, No.
1, pp. 56–70, Feb. 1997.
[16] M. Grossglauser and D. N. C. Tse, “A framework for robust measurement-based admis-sion control,” IEEE/ACM Trans. Networking, Vol. 7, No. 3, pp. 293–309, June 1999.
[17] A. R. Bonde and S. Ghosh, “A comparative study of fuzzy versus ‘fixed’ thresholds for robust queue management in cell-switching networks,” IEEE/ACM Trans. Networking, Vol. 2, No. 4, pp. 337–344, Aug. 1994.
[18] R. G. Cheng and C. J. Chang, “Design of a fuzzy traffic controller for ATM networks,”
IEEE/ACM Trans. Networking, Vol. 4, No. 3, pp. 460–469, June 1996.
[19] I. W. Habib, A. A. Tarraf, and T. N. Saadawi, “Intelligent traffic control for ATM broadband networks,” IEEE Comm. Mag., Vol. 33, No. 10, pp. 76–85, Oct. 1995.
[20] A. Hiramatsu, “ATM communications network control by neural networks,” IEEE Trans. Neural Networks, Vol. 1, No. 1, pp. 122–130, Mar. 1990.
[21] R. J. T. Morris and B. Samadi, “Neural network control of communications systems,”
IEEE Trans. Neural Networks, Vol. 5, No. 4, pp. 639–650, July 1994.
[22] S. A. Youssef, I. W. Habib and T. N. Saadawi, “A neurocomputing controller for band-width allocation in ATM networks,” IEEE J. Select. Areas Commun., Vol. 15, No. 2, pp. 191–199, Feb. 1997.
[23] R. G. Cheng and C. J. Chang, “Neural network connection admission control for ATM networks,” IEE Proceedings Communications, Vol. 144, No. 2, pp. 93–98, Apr. 1997.
[24] B. Kosko, Neural Networks and Fuzzy Systems, Prentice–Hall, 1992.
[25] C. T. Lin, C. S. G. Lee, “Neural-network-based fuzzy logic control and decision system,”
IEEE Trans. Computers, Vol. 40, No. 12, pp. 1320–1336, Dec. 1991.
[26] S. T. Welstead, Neural Network and Fuzzy Logic Applications in C++, John Wiley &
Sons Inc., 1994.
[27] Y. Lin and G. A. Cunningham III, “A new approach to fuzzy-neural system modeling,”
IEEE Tran. Fuzzy Systems, Vol. 3, No. 2, pp. 190–198, May 1995.
[28] N. Yin, S. Q. Li and T. E. Stern, “Congestion control for packet voice by selective packet discarding,” IEEE Tran. Commun., Vol. 38, No. 5, pp. 674–683, May 1990.
[29] D. E. Goldberg, Genetic Algorithms, Addison–Wesley, 1989.
[30] D. L. Gall, “MPEG: A video compression standard for multimedia applications,” Com-mun. of the ACM , Vol. 34, No. 4, pp. 46–58, Apr. 1991.
[31] “Traffic control and congestion control in B-ISDN,” ITU-T Recommendation I.371, Geneva, May 1996.
[32] “Traffic management specification: Version 4.0,” The ATM Forum Technical Commit-tee, Apr. 1996.
[33] S. Q. Li and C. L. Hwang, “Queue response to input correlation functions: continuous spectral analysis,” IEEE/ACM Trans. Networking, Vol. 1, No. 6, pp. 678–692, Dec.
1993.
[34] H. D. Sheng and S. Q. Li, “Spectral analysis of packet loss rate at a statistical multiplexer for multimedia services,” IEEE/ACM Trans. Networking, Vol. 2, No. 1, pp. 53–65, Feb.
1994.
[35] C. W. Therrien, Discrete random signals and statistical signal processing, Prentice–Hall, New Jersey, 1992.
[36] C. J. Chang, C. H. Lin, D. S. Guan and R. G. Cheng, “Design of a power-spectrum-based ATM connection admission controller for multimedia communications,” IEEE Trans. Industrial Electronics, Vol. 45, No. 1, pp. 52–59, Feb. 1998.
[37] C. J. Chang, H. M. Chi, and R. G. Cheng, “A power-spectrum based connection ad-mission control for ATM networks,” Proc. IEEE ICC ’96, Vol. 2, pp. 637–641, June 1996.
[38] C. J. Chang, S. Y. Lin, R. G. Cheng and Y. R. Shiue, “PSD-based neural-net connection admission control,” Proc. IEEE INFOCOM ’97 , Vol. 3, pp. 955–962, April 1997.
[39] C. J. Chang, S. Y. Lin, Y. R. Shiue and R. G. Cheng, “A power-spectrum based neural fuzzy connection admission mechanism for ATM networks,” Proc. IEEE ICC ’97 , Vol.
3, pp. 1709–1713, June 1997.
[40] E. P. Rathgeb, “Modeling and performance comparison of policing mechanism for ATM networks,” IEEE J. Select. Areas Commun., Vol. 9, No. 3, pp. 325–334, Apr. 1991.
[41] L. Dittmann, S. B. Jacobsen and K. Moth, “Flow enforcement algorithms for ATM networks,” IEEE J. Select. Areas Commun., Vol. 9, No. 3, pp. 343–350, Apr. 1991.
[42] S. Shioda and H. Saito, “Satisfying QoS standard with combined strategy for CAC and UPC,” Proc. ICC ’95, Vol. 2, pp. 965–969, June 1995.
[43] M. Butto, E. Cavallero and A. Tonietti, “Effectiveness of the leaky bucket policing mechanism in ATM networks,” IEEE J. Select. Areas Commun., Vol.9, No.3, pp. 335–
342, Apr. 1991.
[44] T. D. Ndousse, “Fuzzy neural control of voice cells in ATM networks,” IEEE J. Select.
Areas Commun., Vol. 12, No. 9, pp. 1488–1494, Dec. 1994.
[45] A. A. Tarraf, I. W. Habib and T. N. Saadawi, “A novel neural network traffic enforce-ment mechanism for ATM networks,” IEEE J. Select. Areas Commun., Vol. 12, No. 6, pp. 1088–1096, Aug. 1994.
[46] V. Catania, G. Ficili, S. Palazzo and D. Panno, “A comparative analysis of fuzzy versus conventional policing mechanisms for ATM networks,” IEEE/ACM Trans. Networking, Vol. 4, No. 3, pp. 449-459, June 1996.
[47] R. G. Garroppo, S. Giordano, S. Miduri and F. Russo, “A prediction based UPC mech-anism for VBR video traffic,” Proc. ICC ’98, Vol. 2, pp. 1119–1123, June 1998.
[48] F. Guillemin, C. Rosenberg and J. Mignault, “On characterizing an ATM source via the sustainable cell rate traffic descriptor,” Proc. INFOCOM ’95, Vol. 3, pp. 1129–1136, April 1995.
[49] R. Braden, D. Clark and S. Shenker, “Integrated services in the Internet architecture:
an overview,” RFC 1633 , June 1994.
[50] S. Blake, D. Black, M. Carlson, E. Davies, Z. Wang and W. Weiss, “An architecture of differentiated services,” RFC 2475 , Dec. 1998.
[51] X. Xiao and L. M. Ni, “Internet QoS: a big picture,” IEEE Magazine on Network , Vol.
13, No. 2, pp. 8–18 Mar./Apr. 1999.
[52] J. Heinanen and R. Guerin, “A single rate three color marker,” RFC 2697 , Sep. 1999.
[53] J. Heinanen and R. Guerin, “A two rate three color marker,” RFC 2698 , Sep. 1999.
[54] W. Fang, N. Seddigh and B. Nandy, “A time sliding window three color marker,” RFC 2859 , June 2000.
[55] F. Wang and P. Mohapatra, “A random early demotion and promotion marker for assure service,” IEEE Journal of Selected Areas in Communications, Vol. 18, No. 12.
pp. 2640–2650, Dec. 2000.
[56] D. D. Clark and Wenjia Fang, “Explicit allocation of best-effort packet delivery service,”
IEEE/ACM Transactions on Networking, Vol. 6, No. 4, pp. 362–373, August 1998.
[57] C. J. Chang, Y. H. Cheng and L. F. Lin, “The traffic conditioner with promotion and fairness guarantee schemes for DiffServ networks,” Proc. of ICC 2003 , Vol. 1, pp. 238–
242, May 2003.
[58] J. M. Jaffe, “Bottleneck flow control,” IEEE Transactions on Communications, Vol. 29, No. 7, pp. 954–962, July 1981.
[59] R. Jain, The art of computer systems performance analysis, John Wiley & Sons Inc., 1991.
[60] S. Floyd and V. Jacobson, “On traffic phase effects in packet-switched gateways,” In-ternetworking: Research and Experience, Vol. 3, No. 3, pp. 115–156, Sep. 1992.
[61] S. Haykin and L. Li, “Nonlinear adaptive prediction of nonstationary signals,” IEEE Trans. Signal Processing, Vol. 43, No. 2, pp. 526–535, Feb. 1995.
[62] J. Baltersee and J. A. Chambers, “Nonlinear adaptive prediction of speech signals using a pipelined recurrent neural network,” IEEE Trans. Signal Processing, Vol. 46, No. 8, pp. 2207–2216, Aug. 1998.
Vita
Li-Fong Lin was born in Tainan, Taiwan, R.O.C., on June 10, 1974. He received B.E.
degree in communication engineering from the Department of Communication Engineering, National Chiao-Tung University, Taiwan, in 1996. Currently, he is a candidate and work-ing toward the Ph.D degree in the area of communication engineerwork-ing, in the Institute of Communication Engineering, National Chiao-Tung University. His current research inter-ests include performance analysis, traffic control over multimedia high-speed networks, and intelligent techniques involving fuzzy logic, neural networks and neural fuzzy systems. The principle considerations of his Ph.D dissertation are the analysis, simulation and optimal design of traffic control schemes in multimedia high-speed networks.