PROJECT TITLE :

OPTOS: A Strategy of Online Pre-Filtering Task Offloading System in Vehicular Ad Hoc Networks

ABSTRACT:

The more advanced services that are offered by vehicle ad hoc networks, also known as VANETs, frequently require individual vehicles to perform complicated computing tasks that they might not be able to finish within the required delay limit. It is common practice to improve service performance in VANET by offloading tasks to road side units (RSUs). Despite this, RSUs may not always have sufficient resources to handle all task offloading requests. Offloading tasks to other vehicles as an alternative to RSU-based task offloading presents itself as a potentially fruitful option in light of the growing amount of computing capacity that is available on vehicles. However, vehicle-to-vehicle task offloading in VANET faces some new challenges that have not been fully addressed. One of these new challenges is the degraded delay performance that is caused by vehicle mobility. In order to find a solution to this issue, we have come up with the idea of developing an Online Pre-filtering Task Offloading System (OPTOS) that is able to reduce the negative effect that vehicle mobility has on the performance of task offloading. OPTOS is comprised of a process that selects candidate vehicles for hosting offloaded tasks and an HGSA algorithm that assigns tasks to vehicles for the purpose of minimizing the delay in the completion of tasks while simultaneously balancing the utilization of the computing capacities on various vehicles. In order to evaluate the effectiveness of the proposed OPTOS, we have carried out extensive experiments making use of a dataset based on the real world. The results that were obtained indicate that OPTOS is effective for reducing the delay in the completion of tasks and increasing the rate at which tasks are completed successfully in a variety of VANET scenarios that involve varying degrees of vehicle mobility.