Research On Computation Offloading In Vehicular Edge Computing

This dissertation focuses on addressing the challenges related to modern vehicular applications and the complexity of vehicular edge computing networks(VECNs).The main objective is to optimize computation offloading within VECNs in terms of cost,delays,and energy consumption.VEC enhances the computational capabilities of vehicles,allowing them to process computation-intensive and delay-sensitive applications at the network edge,resulting in reduced latency and energy consumption.However,the growing demands of modern vehicular applications,such as object detection,self-driving,gaming,and virtual reality,pose challenges in meeting communication and computation requirements.Designing an efficient task offloading scheme for the highly mobile and dynamic vehicular environment is a significant challenge.The dissertation aims to overcome these challenges and optimize computation offloading in VECNs.In the first contribution,our research focuses on the study of cost optimized multi-access edge computing(MEC)based vehicular edge computing networks(VECNs).To address the cost and delay minimization in VECN,we propose a mobility,contact,and computational load-aware(MCLA)task offloading scheme specifically designed for heterogeneous VECN.This scheme takes into account the mobility headway,communication contact,and computational load of vehicles to optimize the task offloading process.To optimize the performance,the MCLA scheme integrates the Mode-1 and Mode-2 of the 5G-NR-V2X standard,along with mmWave communications.The MCLA scheme provides an opportunistic switching mechanism between these modes and heterogeneous radio access technologies(RATs)to reduce communication delays and costs.Moreover,the MCLA scheme leverages public vehicles(i.e.,public buses),in proximity by using their computational power to manage computational latency and cost.Furthermore,it also considers the shareable computations from passengers’ mobile equipment within the public vehicle to improve the computation capacity of the public vehicles.Extensive evaluations and numerical results show that the proposed MCLA scheme significantly improves the task turnover ratio by 4%-15%with 4.7%-29.8%lower transmission and computation costs.The second contribution of this research focuses on optimizing latency and addressing the associated constraints to facilitate the execution of complex applications in VECNs.In this regard,a DRL-based mobility,contact,and load aware cooperative task offloading(DCTO)scheme is developed.DCTO is designed for both cellular and mm Wave radio access technologies(RATs),and both binary and partial offloading mechanisms.DCTO targets delay minimization by opportunistically switching RATs and offloading mechanisms.We consider relative efficacy and neutrality factors as key performance indicators and use them to derive the DRL agent’s reward function.Extensive evaluations demonstrate that the DCTO scheme exhibits a substantial enhancement in task success rate,with an increase from 2.61%to 21.34%.It also improves the efficacy factor from 1.38 to 3.52 and reduces the neutrality factor from 4.99 to 0.76.Furthermore,the average task processing time is reduced by a range of 3.77%to 24.15%.Additionally,the DCTO scheme outperforms the other evaluated schemes in terms of reward and TFPS ratio.The third contribution of our research focuses on achieving energy-efficient and delay-optimized access to the VEC server while addressing the associated constraints and challenges.To overcome the associated challenges,reconfigurable intelligent surfaces(RIS)can play an important role in 6G vehicular networks.With RIS,networks can provide better connectivity,increased data rate and energy efficient access.In this regard,we utilize zero-energy RIS(ze-RIS)to aid vehicular computation offloading while maximizing the energy and time savings while meeting the task and environmental constraints.We formulate an optimization problem that jointly optimizes power control and computation offloading mechanisms.To solve this problem,we convert it into a Markov decision process(MDP)and propose an efficient a DRL-driven ze-RISassisted energy efficient task offloading(DREEO)scheme.DREEO employs a hybrid binary and partial offloading mechanism,allowing vehicles to connect to the VEC server either through direct vehicle-RSU-VEC links or vehicle-RISRSU-VEC links.The DREEO scheme intelligently switches communication links and offloading mechanisms to make efficient decisions that save energy and time.To evaluate the performance of the DREEO scheme,we utilize an efficiency factor as a performance indicator and reward function for our DRL agent.This reward function takes into account two key indicators i.e.,saved energy,and saved time.Through extensive evaluations,DREEO scheme shown an increase in task success rate from 2.13%to 7.36%and has improved the efficiency factor from 21.97 to 51.27.Furthermore,compared to other evaluated schemes,the DREEO scheme consistently outperforms them in terms of reward and the TFPS ratio,the DRL properties.Altogether,this dissertation focuses on enhancing task offloading efficiency in vehicular edge computing(VEC).It introduces three key contributions:a stable and cost-efficient task offloading scheme called mobility,contact,and computational load-aware(MCLA),a DRL-based mobility,contact,and loadaware cooperative task offloading(DCTO)scheme for latency optimization,and a DRL-driven ze-RIS-assisted energy-efficient task offloading(DREEO)scheme.The effectiveness of the proposed approaches is validated through extensive evaluations and numerical results.This dissertation makes significant contributions to the advancement of vehicular edge computing networks by enhancing quality of service,reducing costs,minimizing delays,improving energy efficiency,and enhancing task offloading decision-making capabilities.