https://doi.org/10.1140/epjp/s13360-024-05312-x
Regular Article
Chaotic characteristic analysis and prediction of bottleneck-delay time series under the Internet macro-topology
Faculty of Information, Liaoning University, Shenyang, China
Received:
30
March
2024
Accepted:
25
May
2024
Published online:
6
June
2024
Internet has evolved into a nonlinear and open system that operates far from equilibrium. Aiming at the issue that the network delay is usually in an uncertain state, we primarily construct the Internet macro-topology based on complex network theory. On this basis, we statistically calculate ratios of bottleneck-delays generated on valid paths with different hops and find that the monthly rates have exceeded 50%, which indicates that bottleneck-delay occurs universally. Furthermore, bottleneck-delay time series exhibits stochastic oscillation characteristics throughout the evolution process. Thus, starting from these properties, chaos theory is introduced to analyze its evolution behavior. We use the phase space reconstruction technique and the G-P algorithm to reconstruct the bottleneck-delay time series in phase space. Then, we obtain its chaotic attractor saturation correlation dimension is the fractional dimension and the largest Lyapunov exponent is greater than 0, which confirms that its evolution process has chaotic characteristics., it demonstrates that bottleneck-delay time series has chaotic characteristic. Finally, we propose the chaos-RBF neural network that is adopted the radial basic function (RBF) neural network combining with chaos theory to predict the bottleneck-delay time series. Through error statistics and comparative analysis, the experimental results demonstrate that the chaos-RBF neural network has high prediction accuracy and can better reflect the changing trend of the bottleneck-delay. By calculating the largest Lyapunov exponent, there will be better prediction effects within 5 months in the future.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to Società Italiana di Fisica and Springer-Verlag GmbH Germany, part of Springer Nature 2024. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
