EALSO: joint energy-aware and latency-sensitive task offloading for artificial Intelligence of Things in vehicular fog computing.

Artificial Intelligence of Things (AIoT) is an emerging field within the Internet of Things (IoT) that caters to the growing demand for intelligent computing and communications. Vehicular fog computing (VFC) with AIoT enhance the processing speed of intelligent services. However, this integration also brings about significant energy wastage and high latency problems due to the large amount of data generated by intelligent tasks. To address these issues, our paper introduces EALSO, a novel approach for energy-aware and latency-sensitive task offloading in VFC. EALSO takes into account the quality of service requirements of tasks by mapping them to specific latency constraints and adjustable task priorities. To optimize the energy consumption of VFC networks while ensuring the quality of task offloading, we propose a two-stage hybrid heuristic energy consumption optimization (HHECO) algorithm. In the first stage, a stable matching algorithm is employed for fog vehicle selection. In the second stage, the heuristic algorithm is guided to reduce interference using decision tree techniques and intelligently allocate transmission power to minimize energy consumption and total network latency. We further validate our approach through real-world scenario simulations, demonstrating its superiority over state-of-the-art methods in terms of both energy consumption and latency. [ABSTRACT FROM AUTHOR]