Smoother traffic flows and safer roads could be around the corner with Intelligent Transportation Systems, or ITS, a technology that uses the Internet of Things (IoT) to gather and provide data on traffic speeds, car distances and potential hazards to drivers. While this technology has existed for some time, current deployments are costly and time consuming. To address these constraints, researchers have proposed a 5G-based software-defined network architecture that could reduce the expenses and time associated with ITS deployment.

In the ITS concept, all vehicles on the road would be in constant communication with other vehicles and with the infrastructure. This requires onboard equipment and roadside units (RSUs) to provide directions and alerts to drivers.

The new system minimizes ITS deployment costs by using an architecture incorporated with both 5G –mobile communication that uses less energy and offers expanded system capacity – and software defined-networking (SDN). SDN is a cloud computing architecture where a control plane is physically separated from a forwarding plane that sends off any commands the control plane provides.

With the new ITS architecture, a 5G network gathers and transmits data while the SDN processes data collected by the system. 5G provides more bandwidth and higher data rates to handle the large amounts of data generated by the ITS, and the SDN architecture provides flexibility and easier data management.

The 5G network involves three function layers: the convergence, relay and sensing layer. The core network of the system makes up the convergence layer, and its job is to send out and process data transmitted by a 5G-enabled Multiple-In Multiple Out (MIMO) Network that can handle transmitting multiple data packets at one time.

The relay layer’s role is to assist in communication to individual vehicles in the ITS. Multiple interconnected RSUs disseminate data between the sensing and convergence layers using device-to-device communications. The sensing layer gathers data through IoT-based devices embedded into vehicles as well as IoT sensors dispersed throughout smart cities. With this data, the sensing layer can provide traffic information to drivers after the core network processes data.

ITS SDN 5G Figure 1

Figure 1: Proposed 5G Network Architecture

Within the 5G architecture, the SDN-core network is used to process the data aggregated by the IoT sensors. The SDN-core network uses a SDN controller to optimize data collection and transmission. The controller employs a traffic engineering method to determine the priority of routers in the network based on traffic congestion. This allows the network to provide directions and vehicle positions in real time.

To process data, the SDN core network uses an algorithm within the Hadoop data system to more efficiently process large amounts of data by adjusting network node use if certain nodes aren’t needed. The SDN is also programmed to classify traffic events and generate decisions after filtering out unnecessary data.

“The main purpose of our architecture is to provide constant connectivity between vehicles and the SDN controller,” said Syed Hassan Ahmed, assistant professor, Georgia Southern University. “Big data always requires a large amount of time and processing power, which makes the idea of a real-time ITS cost prohibitive. Our novel architecture provides low-cost implementation with high bandwidth and less end-to-end delay.”

Figure 2 ITS SDN 5G

Figure 2: Processing time in the simulation

To test their system, the researchers ran simulations using C programming language to compare their architecture’s performance against RMERS, an existing ITS architecture. The results showed their system produced more traffic data than the other system, and revealed the researchers’ system consistently required less processing time for queries than RMERS.

Following these positive results, the team wants to move their proposed architecture to real-time implementation. The researchers are also exploring if they can incorporate big data analytics to their system by adding deep learning algorithms to the 5G architecture.

While more development is needed, this research is a major step toward making the ITS more efficient and economical. Soon these developments will enable ITS to create safer and more efficient roads and highways.

For more information on ITS, visit the IEEE Xplore Digital Library.