Abstract : 5G spectral efficiency requirements foresee net- work densification as a potential solution to improve capac- ity and throughput to target next-generation wireless net- works (NGWNs). This is achieved by shrinking the footprint of base stations (BSs), effective frequency reuse, and dynamic use of shared resources between users. However, such a deployment results in unnecessary handovers (HOs) due to the cell size decrements, and limited sojourn time on a high train mobility. In particular, when a train speedily passes through the BS radio coverage footprints, frequent HO rate may result in serious communication interruption impacting quality of service (QoS). This paper proposes a novel context-aware HO skipping that relies on passenger mobility, trains trajectory, travelling time and frequency, network load and signal to interference and noise ratio (SINR) data. We have modelled passenger traffic flows in cardinal directions i.e, north, east, west, and south (NEWS), in a novel framework that employs realistic Poisson point process (PPP) for real-time mobility patterns to support mo- bile networks. Spatio-temporal simulations leveraging NEWS mobility prediction model with machine learning (ML) where support vector machine (SVM) shows an accuracy of 94.51%. ML-driven mobility prediction results integrate into our proposed scheme that shows comparable coverage probability, and average throughput to the no skipping case, while significantly reducing HO costs.
Index terms : 6G, artificial intelligence, context-aware, HO skipping, machine learning, mobility prediction, optimization, smart city planning.