A Complementary Approach to Centralized Task Offloading Algorithms in Vehicular Ad-hoc Networks (VANETs)
Jian Gao, Yuxuan Qi, Wenkai Pan, Yuchen Ma, Xinyi Gong · International Core Journal of Engineering · 2020
Autonomous driving and other advanced applications in our current transportation system have significantly added to the demand of computational power on the road. As industrial research suggested, cloud computing will play an important role in solving the tasks that are generated but unable to be solved by modern vehicles. However, in an urban area, cellular base stations may fail to deliver crucial information due to the existing congestion in wireless channels caused by the large amount of LTE/5G-enabled devices. Therefore, in order to keep a reliable connection between the task generator and the computing server, a scheme of edge computing is introduced to Intelligent Transportation Systems (ITS) in which further optimizations can be made. For example, in the centralized approach of vehicular edge computing, the load of tasks can be balanced among a number of nearby Road Sign Units (RSUs) the Approximate Load Balancing Task Offloading Algorithm (ALBTOA) which is essentially based on game theory. Similarly, we aim to developing a foolproof algorithm to balance the loads effectively from a distributive perspective when data collected is not as much and every vehicle has to decide which server to offload to.