SOFE: SDN-based Task Offloading Technique for Fog-IoEV Network
This paper proposes SOFE, an SDN-based task offloading technique for Fog-IoEV networks that utilizes a binary linear programming model to optimize task assignment to fog nodes or parked vehicles based on load, mobility, energy, and deadlines, thereby reducing response time and extending electric vehicle lifespans compared to existing methods.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine a city where every car is not just a vehicle, but a moving computer, constantly gathering data about traffic, weather, and road conditions. These electric cars, equipped with sensors and cameras, generate a flood of digital requests that need to be solved instantly to keep travel safe and smooth. However, the batteries and processors inside these cars are limited; they cannot handle every single task on their own without running out of power or slowing down. To solve this, engineers have turned to "fog computing," a system that pushes processing power closer to the cars, using roadside equipment and even other vehicles to help out. But a new problem has emerged: if the nearest helper is too busy, or if a parked car helping out runs its battery flat, the whole system can fail, causing dangerous delays or leaving cars stranded.
Researchers have developed a new method to manage this delicate balance, specifically designed for networks of electric vehicles. The team, led by Ahmed Jawad Kadhim Al-Shaibany and colleagues, created a system called SOFE. This approach acts like a central traffic controller for digital tasks, deciding exactly which computer should handle a request and when. Instead of simply sending a task to the closest available machine, the system looks at the bigger picture. It considers how much battery power a parked car has left, whether the owner has agreed to let their car share its resources, and how much work the nearby computers are already doing. The goal is to finish tasks quickly without draining the electric vehicles' batteries to the point of failure.
The core of this new system relies on a smart network controller that knows the status of every node in the city. When a moving car generates a task it cannot handle, it first checks if it can do it itself. If not, it sends the request to the nearest roadside computer. If that computer is overloaded, the request goes to the central controller. This controller then scans all available options, including parked electric vehicles in nearby lots. It filters out any parked cars that are low on battery or whose owners have not given permission. It also checks if a potential helper is moving in the same direction as the car that needs help, ensuring that the connection remains stable and fast. The system then calculates the fastest route and assigns the task to the best-suited machine, whether that is a roadside server or a parked car.
To test how well this idea works, the researchers built a detailed simulation using a digital map of Baghdad. They created a virtual environment with thousands of moving electric vehicles, roadside computers, and hundreds of parked cars with varying battery levels. They pitted their new SOFE method against two other existing techniques used in similar networks. The results showed that the new approach was significantly more effective at keeping tasks on schedule. In simulations where the number of parked cars increased, the new method reduced the average time it took to get a response by about 6.7 percent compared to one competitor and nearly 13 percent compared to another. More importantly, it ensured that a higher percentage of tasks were completed before their deadlines expired.
Perhaps the most critical finding concerned the survival of the electric vehicles themselves. In the simulations, the new method drastically reduced the number of cars that ran out of battery and became "dead" or unusable. While the other methods often drained parked cars by asking them to do too much work, the new system strictly limited how many tasks each parked vehicle could handle based on its remaining energy and the owner's permission. This careful management meant that the number of dead electric vehicles dropped by over 80 percent compared to one of the older methods. The study suggests that by treating parked cars as a valuable but fragile resource, and by using a central controller to make smart, energy-aware decisions, cities can keep their electric vehicle networks running efficiently without sacrificing the vehicles' ability to drive home.
The researchers acknowledge that this work was conducted through computer simulations and that real-world conditions might present different challenges. They propose that future work could involve testing these ideas with 5G technology to speed up communication and developing even smarter ways to balance the load. For now, the study offers a clear path forward: a way to harness the idle power of parked electric cars to support the moving ones, ensuring that the network remains fast, reliable, and sustainable for everyone.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.