Stochastic network model for performance analysis of cloud-fog-enabled intelligent vehicular network

Stochastic network model for performance analysis of cloud-fog-enabled intelligent vehicular network

연구 분야: Networking

학회: Cluster Computing

The emergence of the Internet of Things (IoT) is transforming the conventional devices by replacing them with smart and intelligent ones like smart homes, smartphones, and smart vehicles. With the advent of vehicular ad-hoc networks i.e., VANETs, there is a growing need to enhance vehicular resources for leveraging various driving applications that promote safety and ease of driving, and even pave the way for automated driving. More significantly, with the advent of 5G networks, it is extremely important to combine 5G with vehicular networks. Not to mention, cloud computing has evolved as a significant approach to tackle the problem of insufficient resources for running compute-intensive applications and limited storage capacity for backing up data analogous to car apps and roadside structures. Moreover, we are observing the unfolding of fog and edge computing models that enables the features of cloud computing to be available at the network edge. This approach addresses the issue of latency for time-critical applications and reduces network bandwidth consumption by eliminating the need for vehicle requests to travel across the entire network for processing at the cloud level. This study presents a novel architectural framework for fifth-generation VANETs (5G-V), seamlessly incorporating cloud-fog-edge computing paradigms. Moreover, this study evaluates the performance of 5G-VANETs by assessing a range of measures such as throughput, end-to-end delay and utilization. To contemplate these performance indicators, we suggest a queuing network model that illustrates how messages are disseminated throughout the network. We also generate numerical evidence to corroborate the effectiveness of the contemplated model.