Abstract

One of the key enablers in serving the applications requiring stringent latency in 5G networks is fog computing as it is situated closer to the end users. With the technological advancement of vehicles’ on-board units, their computing capabilities are becoming robust, and considering the underutilization of the off-street vehicles, we envision that the off-street vehicles can be an enormously useful computational source for the fog computing. Additionally, clustering the vehicles would be advantageous in order to improve the service availability. As the vehicles become highly connected, trust is needed especially in distributed environments. However, vehicles are made from different manufacturers, and have different platforms, security mechanisms, and varying parking duration. These lead to the unpredictable behavior of the vehicles where quantifying trust value of vehicles would be difficult. A trust-based solution is necessary for task mapping as a task has a set of properties including expected time to complete, and trust requirements that need to be met. However, the existing metrics used for trust evaluation in the vehicular fog computing such as velocity and direction are not applicable in the off-street vehicle fog environments. In this paper, we propose a framework for quantifying the trust value of off-street vehicle fog computing facilities in 5G networks and forming logical clusters of vehicles based on the trust values. This allows tasks to be shared with multiple vehicles in the same cluster that meets the tasks’ trust requirements. Further, we propose a novel task mapping algorithm to increase the vehicle resource utilization and meet the desired trust requirements while maintaining imposed latency requirements of 5G applications. Results obtained using iFogSim simulator demonstrate that the proposed solution increases vehicle resource utilization and reduces task drop noticeably. This paper presents open research issues pertaining to the study to lead...

Highlights

  • As emerging applications require stringent latency, this will eventually force the cellular networks to advance to 5G and beyond 5G (5G/B5G), where 5G/B5G have to serve a wide range of applications in diversified scenarios [1]

  • The value for λf is obtained from (4), and the result in Fig. 9b shows that when λ is doubled from 2 tasks/s to 4 tasks/s, the λf that a logical cluster has to process increases. Another important observation from this figure is that the logical cluster with high trust value tends to process more tasks than those with relatively low trust value (i.e. Cluster 1 has no λf task to process; whereas Cluster 8 processes the highest amount of tasks)

  • We believe that security alone will not be enough to integrate the vehicles as part of a fog computing infrastructure as a fog needs to ensure security to its customers, and its availability

Read more

Summary

INTRODUCTION

As emerging applications require stringent latency, this will eventually force the cellular networks to advance to 5G and beyond 5G (5G/B5G), where 5G/B5G have to serve a wide range of applications in diversified scenarios [1]. A trust-based solution is required in order to meet the Service Level Agreement (SLA) of applications that can only be served by the fog computing facilities (e.g. augmented reality and smart traffic control). As 5G cells are relatively small, taking the moving vehicles into consideration in Vehicular Fog Computing (VFC) can result in frequent handover, incur additional processing overheads, and degrade the service This has convinced us to only consider off-street vehicles as part of fog and in the subsequent sections of the paper, we refer a vehicle that becomes part of fog computing as a v-fog. Motivated by the above observations, we address the need of trust-based solutions in the off-street v-fog environments to support the applications in 5G and make the following contributions in this paper:.

BACKGROUND
TRUST IN VEHICULAR ENVIRONMENTS
PERFORMANCE EVALUATION
OPEN RESEARCH ISSUES
Findings
CONCLUSION
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call