Use of mobile network analytics for application performance design

With the 5G technology, data traffic is going to grow by a factor of 1000, while the number of connected devices is likely going to be two orders of magnitude higher. With smartphones being cornerstone in our daily lives, understanding mobile network performance is critical for providing a superior user experience and, consequently, determining the success of an application. This paper presents a solution that uses the radio parameters measured by a mobile terminal to determine the best Application Protocol (APPP) for a service, so as it could adapt to the varying network conditions. From the training of an inference system with actual Mean Opinion Score (MOS) data, it will be possible to discern which radio Key Performance Indicators (KPIs) are best suited to characterize the state of the network and make the best possible decision. Results show how the decision system based on only three radio KPI is able to determine the user application experience with a success of up to 83%. Thanks to the use of this approach, application developers may fill the gap of knowledge between network KPIs and user experience.