A V2V Task Offloading Decision Algorithm for Multi-RAT Vehicular Networks
Qiaonan Zhu, Jun Gang Zheng · 2024
This paper investigates the vehicle to vehicle (V2V) task offloading problem in a vehicular network with multiple radio access technologies (RATs). The problem is formulated as a mixed integer nonlinear programming problem (MINP) with an objective to minimize the average offloading delay of all offloading tasks in the network by optimizing a set of offloading decisions subject to delay and computing resource constraints. To solve the formulated problem, a log-sum-exp approximation is used to transfer the MINP problem to a combinatorial optimization problem to obtain a probability distribution of all possible offloading decisions. Based on the probability distribution, a Markov-chain based method is used to obtain the state transition probabilities between different offloading decisions. Based on the state transition probabilities, a V2V task offloading decision (V2V-TOD) algorithm is proposed to make offloading decisions via using a learning process consisting of three stages: SV and RAT selection, sub-channel selection, computing resource allocation. Simulation results show that the proposed V2V-TOD algorithm can efficiently reduce the average offloading delay as compared with three benchmark algorithms.