Work

Forecasting Core IP Traffic Using ARIMA-GARCH

Year: 2026

Type: article

Abstract: The rapid expansion of Internet-based services, including cloud computing, video streaming, IoT, and emerging 5G/6G applications, has increased both the volume and complexity of core IP network traffic. Accurate short-term traffic forecasting is essential for capacity planning, congestion control, and quality of service assurance. Core IP traffic exhibits non-stationarity, burstiness, and time-varying variance, which limits the effectiveness of traditional ARIMA models that assume constant residual variance. This paper proposes a hybrid ARIMA–GARCH framework, where ARIMA models the conditional mean and GARCH captures residual volatility and heteroscedasticity.

Source: International Journal of Innovative Science and Research Technology (IJISRT)

DOI: 10.38124/ijisrt/26sep342

Authors: Choukri Benhamed, Slimane Mekaoui

Institution: University of Sciences and Technology Houari Boumediene

Language: English

Topic: Network Traffic and Congestion Control

Subfield: Computer Networks and Communications

Field: Computer Science

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: diamond

OpenAlex ID: W7214104530