Abstract

With the advancement from 4G to 5G networks, mobility management poses a new set of issues in providing uninterrupted service to latency-sensitive applications, such as augmented reality (AR), autonomous driving, and live video streaming. Conventional bandwidth handover techniques primarily emphasize adequate signal strength and congestion information, offering inadequate handover methods to fit the varied and changing demands of these modern applications. The authors introduce a novel handover decision model that incorporates the context of the application, in addition to real-time user mobility patterns and network conditions, so that handover decisions can be made effectively. Using on-device artificial intelligence/machine learning (AI/ML) algorithms, coupled with edge cloud computing, a model of handover predictions will suggest the most appropriate base station for handover across all movements, reducing latency, reducing jitters, and reducing packet losses. Simulated experiments conducted in an urban environment demonstrate that context-aware handover decisions outperformed traditional handover decision methods as users transitioned through bandwidth-intensive applications. Collectively, findings indicate network resource usage was increased, handover efficiency was increased, and the user experience was enhanced, demonstrating the significance of context-aware handover decision frameworks in next-generation mobile networks. In sum, the research advances a foundation for adaptive mobility management decision-making associated with application requirements in 5G and beyond.