[Paper Review] Distributed Vehicular Computing at the Dawn of 5G: a Survey

Ahmad Alhilal, Finley, Benjamin+3 more|arXiv (Cornell University)|Jan 20, 2020

Vehicular Ad Hoc Networks (VANETs)Engineering111 references17 citations

TL;DR

This survey provides a comprehensive analysis of distributed vehicular computing in the 5G era, focusing on edge and cloud computing architectures to enable low-latency, context-aware, and reliable vehicular applications. It integrates 5G, edge computing, big data processing, and emerging 6G visions to address challenges in real-time decision-making, dynamic modeling, and security in intelligent transportation systems.

ABSTRACT

Recent advances in information technology have revolutionized the automotive industry, paving the way for next-generation smart vehicular mobility. Vehicles, roadside units, and other road users can collaborate to deliver novel services and applications. These services and applications require 1) massive volumes of heterogeneous and continuous data to perceive the environment, 2) reliable and low-latency communication networks, 3) real-time data processing that provides decision support under application-specific constraints. Addressing such constraints introduces significant challenges for current communication and computing technologies. Relatedly, the fifth generation of cellular networks (5G) was developed to respond to communication challenges by providing for low-latency, high-reliability, and high bandwidth communications. As a major part of 5G, edge computing allows data offloading and computation at the edge of the network, ensuring low-latency and context-awareness, and 5G efficiency. In this work, we aim at providing a comprehensive overview of the state of research on vehicular computing in the emerging age of 5G and big data.

Motivation & Objective

• To analyze the integration of 5G and edge computing in enabling real-time, low-latency vehicular applications.
• To identify key challenges in data heterogeneity, intermittent connectivity, and dynamic driving environments.
• To explore the role of big data processing, continuous learning, and secure remote computation in vehicular systems.
• To examine the transition from 5G to 6G for future intelligent transportation systems with sub-millisecond latency requirements.
• To assess the feasibility and research gaps in deploying AR windshields, C-V2X, and adaptive models in real-world vehicular environments.

Proposed method

• Survey-based synthesis of state-of-the-art research in vehicular computing, 5G, edge/cloud computing, and big data architectures.
• Categorization of vehicular applications into infotainment, safety-critical, and traffic management based on communication and computation requirements.
• Analysis of vehicular edge computing (VEC) and vehicular cloud computing (VCC) for offloading computation and reducing latency.
• Evaluation of continuous learning techniques (e.g., incremental learning, GDPC) to handle non-stationary driving patterns and avoid catastrophic forgetting.
• Examination of security mechanisms such as data attestation and trusted remote execution for edge-computed results.
• Exploration of augmented reality (AR) windshields and their latency and HCI challenges, including motion-to-photon delay and object selection.

Experimental results

Research questions

• RQ1 How can 5G and edge computing jointly address the low-latency and high-reliability requirements of safety-critical vehicular applications?
• RQ2 What are the key challenges in processing heterogeneous, continuous, and bursty vehicular data streams in real time?
• RQ3 How can machine learning models adapt to dynamic driving patterns without catastrophic forgetting?
• RQ4 What are the architectural and security requirements for trusted remote execution in vehicular edge environments?
• RQ5 What role will 6G play in enabling future applications like the tactile internet with sub-1ms latency in vehicular networks?

Key findings

• 5G-enabled edge computing significantly reduces latency and improves reliability for real-time vehicular applications, especially when combined with C-V2X for safety-critical communication.
• Continuous learning techniques such as incremental learning and Gaussian-based dynamic probabilistic clustering (GDPC) enable models to adapt to changing driving patterns while mitigating catastrophic forgetting.
• AR windshields face critical challenges due to motion-to-photon latency and misalignment, requiring low-latency networking and advanced HCI techniques for usability.
• Current 5G systems fall short of supporting sub-1ms end-to-end latency required for tactile internet use cases, necessitating the development of 6G with enhanced KPIs.
• Secure remote computation via data attestation and trusted execution environments is essential for ensuring correctness and integrity of edge-computed results in vehicular systems.
• Big data processing in vehicular networks demands scalable, context-aware architectures that fuse heterogeneous data sources from vehicles, RSUs, and infrastructure for real-time decision support.