185. MOBILE DATA TRAFFIC ANALYSIS IN VIRTUAL PRIVATE CLOUD: A CASE-STUDY IN OPERATIONAL MOBILE NETWORK

Ana Bogdanovska; Vladimir Atanasovski

doi:10.51466/JEEIT2161185049b

Ana Bogdanovska Faculty of Electrical Engineering and Information Technologies, Ss. Cyril and Methodius University in Skopje, Macedonia
Vladimir Atanasovski Faculty of Electrical Engineering and Information Technologies, Ss. Cyril and Methodius University in Skopje, Macedonia

DOI: https://doi.org/10.51466/JEEIT2161185049b

Keywords: Cloud, virtualization, traffic analytics, log-normal distribution, normal distribution

Abstract

Mobile data traffic analysis is quintessential for mobile telecom operators in order to improve the quality-of-service for its customers. A mobile operator is interested in the usage popularity of services as well as the temporal distribution of the service usage. This enables identification of the volume of incoming and outgoing traffic and aids the operators to design a network that can support expected user activity. The latest networking trends point in the direction of virtualized network resources, thus exhibiting more effective network usage. The design of virtual networks is strongly affected by the pattern of mobile data for certain services. Therefore, a prior and adequate knowledge of the mobile services’ popularity and their usage can significantly improve the user experience on the network. This paper provides empirical measurements and analytical modelling of mobile data traffic using data from an operational nationwide mobile network. The network utilizes the latest trends, i.e., a virtual environment and a private cloud. The paper proposes estimation parameters for the measured traffic that match the distribution of real traffic with the lowest error.