The delay-sensitive applications, such as self-driving, smart transportation, navigation, and augmented reality assistance, can be evolved in vehicular ad-hoc networks (VANETs) using one of the leading paradigms, fog computing (FC). The intelligent vehicles are connected to the roadside infrastructure, such as high power nodes (HPNs) and roadside units (RSUs), also called fog nodes (FNs), for obtaining on-demand services. These FNs possess finite resources and can provide services to limited vehicles. However, when vehicles reach the network spike in demand, the FNs become impuissant in furnishing services in the existing solutions. As a result, there is a significant reduction in the network throughput. Therefore, we propose an efficient resource orchestration (ERO) algorithm to maximize the throughput by reducing the allocated resource blocks (RBs) of FNs. The ERO algorithm partitions the FN coverage region into restricted and non-restricted coverage regions. Then, it coordinates the RBs allocation among FNs by reducing RBs for the vehicles in the non-restricted coverage regions. This reduction is carried out by migrating RBs for offloading upstream services so that the overall occupied capacity of FNs is minimized. ERO constructs the minimum priority queue using the occupied capacity of FNs to perform optimal RBs migration between pairs of FNs. The ERO algorithm is evaluated, and simulation results show that the proposed algorithm performs better in terms of throughput, serviceability, availability, and service capability than existing algorithms.

Vehicular networks use roadside infrastructures, roadside units (RSUs) and high-power nodes (HPNs), as fog nodes (FNs) in intelligent transport systems (ITSs). However, blending fog computing (FC) in ITS can stimulate the network with latency-sensitive applications, such as self-driving, augmented reality assistance and navigation. FNs assign resource blocks (RBs) to vehicles to furnish the services. However, as the vehicles that reach the network grow, FNs’ energy consumption increases. Consequently, FNs become ineffective in delivering services. Therefore, to handle this issue, we present an energy-efficient resource allocation (EERA) algorithm to harmonise RB allocation among FNs, such that the energy utilisation of FNs is reduced. EERA algorithm relocates the assigned RBs of vehicles in overlap coverage amid pairs of FNs, such that the allocated RBs of FNs are minimised. The simulation outcomes show that the proposed algorithm minimises the energy consumption of FNs up to 42.5% on average compared to the existing algorithms.

Emerging technologies, such as the fifth generation (5G) and the Internet of Things (IoT), increase the communication capabilities of components such as smart vehicles in intelligent transportation systems. Consequently, there is a demand for vehicular services to fulfil the purpose of safe driving and comfort in smart transportation and augmented reality assistants. These vehicular services are delay-sensitive tasks and computation-intensive tasks. Hence, these tasks are not ideal for vehicle processing due to stringent deadlines, finite resource constraints and the battery life of vehicles. Therefore, they are handled by offloading into roadside infrastructures (e.g., roadside units or high power nodes), called fog nodes (FNs), for further processing. However, when the delay-sensitive tasks increase in the network during peak time, the processing of such tasks in FNs poses a challenge regarding meeting deadlines and energy consumption. Therefore, we propose an energy and priority-aware scheduling (EPAS) algorithm to handle the delay-sensitive tasks in the overlap coverage areas of fog-enabled vehicular networks (FEVNs) such that the energy consumption of FNs is reduced while meeting deadlines. Task scheduling among FNs is a multiple 0/1 knapsack, a well-known nondeterministic polynomial (NP)-hard problem. Hence, the EPAS is a greedy-based sub-optimal solution to the task scheduling problem with a finite number of tasks and FNs in FEVNs. The performance of EPAS is evaluated by considering the peak arrival of tasks into the network. The simulation outcomes depict that the EPAS algorithm lowers the FN’s energy consumption compared to benchmark algorithms.

The new computing paradigm, fog computing, enables advanced services, such as navigation, self-driving cars, and augmented reality, by integrating with vehicular networks to provide smart transportation solutions. This emerges fog-empowered vehicular ad-hoc networks (FEVANETs), which enable smart vehicles to communicate with fog nodes (FNs) using association policy, such as signal strength and/or favorite contents. However, a deluge arrival of vehicles to the network can cause a load imbalance among FNs. This impacts the severe reduction in the network service capability and resource utilization efficiency. To address this problem, we propose an algorithm, dynamic resource management (DRM), for assigning the resources of FNs to smart vehicles by migrating services among FNs. The problem is formulated as integer linear programming (ILP) and determines its NP-hardness by reducing it from Seminar Assignment Problem. A polynomial-time algorithm is presented by formulating the problem as a graph in which vertices represent the FNs and edges represent the vehicles present in the overlapped region of the pairs of FNs. The proposed algorithm considers the set of vehicles that are in overlapped coverage regions of FNs and communicates with those corresponding FNs. Then it migrates the resource blocks (RBs) of the set of vehicles between pairs of FNs to minimize the allocated RBs. The DRM is simulated extensively, and the simulation outcomes show that the DRM enhances service capability, serviceability, availability, throughput, and resource utilization efficiency compared to the four existing algorithms.