Dual Connectivity Management in 5G Mobile Internet Infrastructures

The newly developed mobile 5G network is configured to deliver a high-performance interface. Specifically, it facilitates the connection between people, machines, objects, and devices. With its breakthrough capabilities, including higher data rates, massive capacity, and ultra-low latency, new user experiences such as augmented and virtual reality are now possible. Furthermore, new service domains, such as IoT connectivity, are made possible. An essential characteristic of 5G deployment is dual connectivity (DC), which provides additional support for seamless integration with already existing 4G core systems. This paper introduces an MM algorithm that facilitates a smooth transition for users between 4G interfaces and 5G interfaces (and vice versa). Our approach leverages DC because it reduces handover interruptions compared to traditional hard handovers. The proposed MM scheme implements a dynamic data-splitting approach between 4G and 5G RATs depending on the type of application requirement. To validate our proposed solution, we utilize dual connectivity to propel strategy-driven data splitting for mobile users in different market verticals. The system is modeled as a Markov Decision Process (MDP) where a controller employs reward-based selections to break non-deterministic choices. We incorporate a probabilistic model-checking framework to validate our solution, which is tested through various scenarios to assess the functionality of the model-checking tool.