Abstract
This publication describes systems and techniques to manage network handovers between terrestrial networks and non-terrestrial networks using a machine-learned model and wearable device augmentation. In hybrid connectivity environments, electronic devices can experience data stalls or unnecessary battery drain when relying solely on instantaneous signal strength for handover decisions. To enhance connectivity, the described systems utilize a machine-learned model to predict a superior handover time based on a probabilistic function. The model processes inputs that include signal trends, network performance metrics, satellite gap data, and battery state. A wearable device is integrated to provide physical and contextual augmentation. The wearable device acts as a secondary receiver or relay to mitigate line-of-sight obstructions, captures human context data via inertial sensors to inform the machine-learned model, and delivers haptic feedback to guide a user in orienting the device for optimal satellite reception. By evaluating these factors, the system can preemptively establish non-terrestrial connections, facilitate efficient network scanning, and mitigate data stalls when signal strength is high, but data throughput is low.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
J., Karthick M.; Cha, Youngtae; Mallampati, Aishwarya; Sasindran, Sooraj; Nguyen, Thomas; Paul, Soumyadeep; and Choi, Hakjun, "Machine-Learned Probabilistic Handover between Terrestrial and Non-Terrestrial Networks Utilizing Wearable Device Augmentation", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/10398