PROJECT TITLE :

Partial Computation Offloading and Adaptive Task Scheduling for 5G-enabled Vehicular Networks

ABSTRACT:

In order to pique the interest of prospective users in the emerging 5G-enabled vehicular networks, a wide variety of cutting-edge mobile applications are currently being developed. Despite the fact that computation offloading and task scheduling have been the subject of extensive research, it is still rather difficult to determine the optimal offloading ratio and successfully carry out adaptive task scheduling in networks with a high degree of dynamism. In addition, because vehicular users are rational and self-interested, the scheduling policy created by the network operator may be broken, as their goal is to maximize their own profits. We present POETS, an effective algorithm for partial computation offloading and adaptive task scheduling that takes into account the incentive compatibility and individual rationality of vehicular users. The goal of this algorithm is to maximize the overall system-wide profit. In particular, a two-sided matching algorithm is proposed as the first step toward deriving the optimal transmission scheduling discipline. After that, the offloading ratio of vehicular users can be determined through convex optimization, and this can be done so without any knowledge of the other users. In addition, a non-cooperative game is built to derive the payoff of vehicular users so that an equilibrium can be reached between users and the network operator. This equilibrium is the goal of the construction of the game. The efficacy of our proposed solution has been demonstrated through both theoretical investigations and performance evaluations that are derived from actual traces of taxis from the real world.