مينا خوش بزم فريماني

Smart vehicles have a rising demand for computation resources and recently vehicular edge computing is realized to be an effective workaround. Edge servers deployed in the roadside units are capable of accomplishing tasks beyond the capacity which is embedded inside the vehicles. However, the main challenge is to carefully select the tasks to be offloaded considering the deadlines and in order to reduce energy consumption while delivering a good performance. In this thesis, we consider e a vehicle edge computing network architecture in which multiple cars are moving at none-constant speeds and produce tasks at each time slot. Then, we propose a task offloading algorithm with regard to the vehicle's direction based on Rainbow, a deep Q-learning algorithm combining several independent improvements to the DQN algorithm with the aim of overcoming the limitations of traditional DQN, to seek an optimal offloading policy by effectively incorporating the computation resources of edge servers in order to jointly minimize average delay and energy consumption. Real-world traffic data is used to eva‎luate the performance of the proposed approach compared to other DQN algorithms namely DQN, Double-DQN, and Deep Recurrent Q-learning. Results of the experiments show an average reduction of 18% and 15% in energy consumption and delay, respectively, when using the Rainbow algorithm compared to the state-of-the-art. Moreover, the stability and convergence of the learning process have visibly improved by adopting the Rainbow algorithm. Keywords: Task Offloading – Mobile Edge Computing – Vehicular Networks – low latency

بارسپاري وظايف – محاسبات لبه‌اي سيار – شبكه‌هاي بين خودرويي – تأخير كم

Task Offloading – Mobile Edge Computing – Vehicular Networks – low latency

mina khoshbazm farimani

reza entezari-maleki