Comparative Analysis of Forecasting Methods for 5G Using Real Time Mobile Network Data

Dr. Anupriya

Abstract: The increasing complexity and heterogeneity of modern network traffic present significant challenges for accurate traffic classification in Software-Defined Networking (SDN) environments. Conventional classification methods repetitively struggle to comply with expeditiously evolving traffic behaviors, resulting in suboptimal resource management. To address this issue, an enhanced AGBFM (Adaptive Gradient Based Forecasting Model) and EWOFM (Enhanced Whale Optimization Based Forecasting Model) is proposed for efficient and accurate SDMN traffic classification. The incorporation of optimization algorithms improves the efficiency of the standard forecasting models, making it suitable for dynamic SDMN environments. Experimental results determine that the proposed models accomplishes a classification accuracy of 96.60%, surpassing LSSVM-PSO, LSSVM-ACO and LSSVM-WOA using the similar dataset scenarios. These findings show that AGBFM and EWOFM offer a lightweight yet effective solution for real-time traffic forecasting in SDMN.

Keywords: Forecasting, Adaptive, Gradient, Optimization, traffic