Enhancing Mobile Network Performance Through ORAN-Integrated UAV-Based Mobility Management

The open radio access network (O-RAN) architecture promotes the evolution of the RAN through the disaggregation of functions, the adoption of open interfaces, and the instantiation of a hierarchical closed-loop control architecture managed by the RAN intelligent controllers (RICs). This architecture enables interoperability, flexibility, and intelligence between different network elements and the integration of various artificial intelligence (AI) and machine learning (ML) applications into the network. One of the key benefits of O-RAN is the continuous monitoring of quality-of-service (QoS) metrics and network resources, which can be used to make optimized decisions about resource allocation, management, and scaling. One of the network features that can benefit from this approach is the use of unmanned aerial vehicles as flying base stations (UAV-BSs), which can be deployed to provide wireless connectivity to ground devices, complementing the terrestrial infrastructure, such as in high-demand scenarios, infrastructure failures, or disasters. However, deploying and controlling UAV-BSs are complex challenges involving aspects such as positioning, power consumption, interference, safety, and mobility. In this context, this paper proposes an approacH that Adjusts drones as flying Base statIons using weighTed k-meAns for the oraN archiTecture (HABITANT), a new approach for the O-RANs architecture, consisting of the use of a positioning algorithm running at an extensible application (xApp) to calculate optimized locations for UAV-BSs. The algorithm takes into account the QoS metrics of user applications, as well as the physical and operational constraints of UAV-BSs. The goal is to provide a more efficient and adaptive solution to the UAV-BS positioning problem, improving the network coverage and the QoS of the users.