Enhanced whale optimization based traffic forecasting for SDMN based traffic

• Published in: ICT express 2021, 7(2), , pp.143-151
• Main Authors: Anupriya, Singhrova, Anita
• Format: Article
• Language: English
• Subjects: Software defined mobile networks; Support Vector Machine; Whale Optimization algorithm; 전자/정보통신공학
• ISSN: 2405-9595; 2405-9595
• DOI: 10.1016/j.icte.2021.05.005

In today’s scenario the number of mobile devices is increasing by leaps and bounds and hence the networks are highly congested. To use the existing networks efficiently, the insight into the amount of traffic beforehand, would be highly beneficial. Mobile traffic forecasting would enable Software Defined Mobile Network (SDMN) to use network resources and bandwidth effectively. Therefore, this paper aims to forecast mobile traffic for SDMN using nature-inspired​ optimization techniques. The data taken is captured using Wireshark. In order to accurately predict network traffic, this paper considers Least Square Support Vector Machine (LSSVM) function estimation based training model with Whale Optimization Algorithm (WOA). The proposed technique increases the probability of global search by optimizing kernel parameters namely, regularization and kernel width. The parameters: Mean Square Error (MSE), True Positive Rate (TPR), True Negative Rate (TNR),​ accuracy and precision are evaluated for the proposed optimization algorithm. The results of the proposed algorithm Least Square Support Vector Machine-Enhanced Whale Optimization Algorithm (LSSVM-EWOA) for traffic forecasting are compared with Least Square Support Vector Machine-Whale Optimization Algorithm (LSSVM-WOA). The accuracy of prediction of the proposed model is 98.10%, which confirmed the effectiveness and the actual use of the proposed model.