Recursive Selection of Hyperexponential Distributions in Approximation of Distributions with "Heavy Tails"
https://doi.org/10.31854/1813-324X-2023-9-2-40-46
Abstract
It is known that many quantities that determine the network characteristics of the functioning of an infocommunication network have probability distributions with "heavy tails", which can have a significant impact on network performance. Models with heavy-tailed distributions tend to be difficult to analyze. The analysis can be simplified by using an algorithm to approximate a heavy-tailed distri-bution by a hyperexponential distribution (a finite mixture of exponentials). The paper presents a algorithm for calculating the parameters of the hyperexponential distribution components, which is based on a recursive selection of parameters. This algorithm allows you to analyze various models of queues, including G/G/1. It is shown that the approach under consideration is applicable to the approxi-mation of monotonically decreasing distributions, including those with a "heavy tail". Examples of approximation of Pareto and Weibull distributions are given.
About the Authors
M. Buranova
Povolzhskiy State University of Telecommunications and Informatics
Russian Federation
Russian Federation
Samara, 443010, Russian Federation
V. Kartashevskiy
Povolzhskiy State University of Telecommunications and Informatics
Russian Federation
Russian Federation
Samara, 443010, Russian Federation
References
1. Kleinrock L. Queueing Systems. Vol. 1: Theory. New York: Wiley-Interscience, 1975. 432 p.
2. Sheluhin O.I., Smolsky S.M., Osin A.V. Self-Similarity and Fractals. Telecommunication Applications. Moscow: Fizmatlit Publ.; 2008. 368 p. (in Russ.)
3. Kotz S., Johnson N.L., Read C.B. Encyclopedia of Statistical Sciences. New York: Wiley; 1988. vol.8. p.352‒357.
4. Keilson J., Machihara F. Hyperexponential waiting time structure in hyperexponential H_K/H_L/1 system. Journal of the Operation Research Society of Japan. 1985;28(3):242‒250. DOI:10.15807/jorsj.28.242
5. Feldmann A., Whitt W. Fitting mixtures of exponentials to long-tail distributions to analyze network performance models. Performance Evaluation. 1998;31(3–4):245‒279. DOI:10.1016/S0166-5316(97)00003-5
6. Buranova M.A., Kartashevskii V.G. Determination of the Parameters of Hyperexponential Distribution by the Method of Recursive Selection. Proceedings of the XXVIIth International Technical Conference on Radar, Navigation, Communications, 28‒
7. September 2021, Voronezh, Russia. Voronezh: Voronezh State University Publ.; 2021. p.43‒54. (in Russ.)
8. Korolyov V.Yu. The EM Algorithm, Its Modifications, and Their Application to the Problem of Separating Mixtures of Probability Distributions. Theoretical Review. Moscow: IPI RAN Publ.; 2007. 94 p. (in Russ.)
9. Buranova M., Ergasheva D., Kartashevskiy V. Using the EM-algorithm to Approximate the Distribution of a Mixture by Hyperexponents. Proceedings of the International Conference on Engineering and Telecommunication, EnT, 20‒21 November 2019, Dolgoprudny, Russia). IEEE; 2019. DOI:10.1109/EnT47717.2019.9030551
10. Tarasov V.N., Kartashevskij I.V. Determination of the average Waiting Time for Requirements in a Managed Mass Service System. Sistemy upravleniya i informaci-onnye tekhnologii. 2014;3(57):92‒96. (in Russ.)
Review
For citations:
Buranova M., Kartashevskiy V. Recursive Selection of Hyperexponential Distributions in Approximation of Distributions with "Heavy Tails". Proceedings of Telecommunication Universities. 2023;9(2):40-46. (In Russ.) https://doi.org/10.31854/1813-324X-2023-9-2-40-46
Views:
378
Email this article 