Computational Offloading for Wearables-Based 6G Fitness Applications

Sports and fitness activities play a vital role in modern society by reducing stress and enhancing overall health. With advancements in semiconductor technologies and communication systems driven by the Internet of Things (IoT), wearable devices have become increasingly compact, capable, and sensor-rich. These devices support monitoring vital signs and track physical activities in real time. However, wearables face significant challenges due to limited computational capacity, low power, and platform heterogeneity, especially when handling data-intensive, real-time processing. To address these issues, this study introduces an efficient task-offloading framework for fitness-oriented wearable devices. The proposed approach employs a convex optimization algorithm with multiple constraints, dynamically adapting to network conditions to optimize offloading decisions. Experimental results demonstrate that the framework enhances execution speed by 15.8% and 7.9%, and reduces energy consumption by 17.6% and 8.8%, compared to local and cloud-based execution methods, respectively.