Mobile traffic flow prediction using intelligent whale optimization algorithm

Anupriya,Singhrova, Anita
DOI: https://doi.org/10.1007/s10515-022-00349-7
IF: 1.677
2022-07-08
Automated Software Engineering
Abstract:The digital growth has triggered explosion of mobile and wireless scenario. This expansion propelled the high demand of wireless capacity especially bandwidth. The increased demand lead to the need of efficient utilization of critical resources in networks. Under such circumstances, software defined mobile network (SDMN) controller has emerged as one of the promising solution for efficient management of resources. As the data flow is not constant throughout, so is the resource requirement. The mobile traffic forecasting helps SDMN controller to allocate resources according to the fluctuating demand of traffic peaks or troughs. Different forecasting algorithms already exist to identify the solution but most of them fail to achieve the global optimum value. This paper motivates to make SDMN mobile network more reliable, congestion free and intelligent decision maker by introducing an intelligent whale optimization algorithm (IWOA) to identify optimal parameters of the forecasting model. The accuracy of the proposed model will improve network efficiency because of dynamic decisions based on forecasting results. The WOA offers slow rate of convergence along iterative process and tends to converge into local optimum. The proposed algorithm is predominantly using chaotic maps, weight factor and convergence factor to estimate and naturally adjust the intrinsic parameters of optimization. Along the iterative cycles, the proposed technique (IWOA) emend the effectiveness of search to reach towards the optimal solution. To illuminate the efficiency of the IWOA in forecasting model, it is compared over two different scenarios of datasets. Additionally, the results show the improved performance of the proposed IWOA in terms of sensitivity (0.02%), accuracy (3.57%), precision (0.05%) and F1-Score (0.04%) with regard to WOA.
computer science, software engineering
What problem does this paper attempt to address?