The Internet of Vehicles (IoV) is attracting many researchers with the emergence of autonomous or smart vehicles. Vehicles on the road are becoming smart objects equipped with lots of sensors and powerful computing and communication capabilities. In the IoV environment, the efficiency of road transportation can be enhanced with the help of cost-effective traffic signal control. Traffic signal controllers control traffic lights based on the number of vehicles waiting for the green light (in short, vehicle queue length). So far, the utilization of video cameras or sensors has been extensively studied as the intelligent means of the vehicle queue length estimation. However, it has the deficiencies like high computing overhead, high installation and maintenance cost, high susceptibility to the surrounding environment, etc. Therefore, in this paper, we propose the vehicular communication-based approach for intelligent traffic signal control in a cost-effective way with low computing overhead and high resilience to environmental obstacles. In the vehicular communication-based approach, traffic signals are efficiently controlled at no extra cost by using the pre-equipped vehicular communication capabilities of IoV. Vehicular communications allow vehicles to send messages to traffic signal controllers (i.e., vehicle-to-infrastructure (V2I) communications) so that they can estimate vehicle queue length based on the collected messages. In our previous work, we have proposed a mechanism that can accomplish the efficiency of vehicular communications without losing the accuracy of traffic signal control. This mechanism gives transmission preference to the vehicles farther away from the traffic signal controller, so that the other vehicles closer to the stop line give up transmissions. In this paper, we propose a new mechanism enhancing the previous mechanism by selecting the vehicles performing V2I communications based on the concept of road sectorization. In the mechanism, only the vehicles within specific areas, called sectors, perform V2I communications to reduce the message transmission overhead. For the performance comparison of our mechanisms, we carry out simulations by using the Veins vehicular network simulation framework and measure the message transmission overhead and the accuracy of the estimated vehicle queue length. Simulation results verify that our vehicular communication-based approach significantly reduces the message transmission overhead without losing the accuracy of the vehicle queue length estimation.
In recent years, we observed the proliferation of cloud data centers (CDCs) and the Internet of Things (IoT). Cloud computing based on CDCs has the drawback of unpredictable response times due to variant delays between service requestors (IoT devices and end devices) and CDCs. This deficiency of cloud computing is especially problematic in providing IoT services with strict timing requirements and as a result, gives birth to fog/edge computing (FEC) whose responsiveness is achieved by placing service images near service requestors. In FEC, the computing nodes located close to service requestors are called fog/edge nodes (FENs). In addition, for an FEN to execute a specific service, it has to be provisioned with the corresponding service image. Most of the previous work on the service provisioning in the FEC environment deals with determining an appropriate FEN satisfying the requirements like delay, CPU and storage from the perspective of one or more service requests. In this paper, we determined how to optimally place service images in consideration of the pre-obtained service demands which may be collected during the prior time interval. The proposed FEC environment is scalable in the sense that the resources of FENs are effectively utilized thanks to the optimal provisioning of services on FENs. We propose two approaches to provision service images on FENs. In order to validate the performance of the proposed mechanisms, intensive simulations were carried out for various service demand scenarios.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.