NSF Award: CNS-1217572
NeTS: Small: Information Dissemination in Vehicular Networks for Reducing Traffic Congestion
With the increment of population, vehicle transportation problems like traffic congestion and accidents are becoming more severe in our society. To enhance the transportation environment, Intelligent Transportation Systems (ITS) have been deployed to bring connectivity to transportation. As a core component of ITS, Vehicular Ad-Hoc Network (VANET) has received considerable attention on information sharing and data delivery services. It is able to offer direct communications between vehicles and between vehicles and roadside units (RSUs). Connected vehicles can send and receive hazard information on the current traffic situation and therefore alert drivers the potential dangerous conditions like icy road or impending collisions. A large number of applications have been proposed based on the advent of DSRC devices for both safety and non-safety purposes. In order to support such variety of applications, we aim at improving the reliability and efficiency of data transmission in VANET.
The major goals of this project are
- To improve the performance of protocols used in vehicular networks
- To integrate vehicular networks with applications like ITS (Intelligent Transportation Systems)
- To develop new applications based on future vehicular networks and ITS infrastructure
We have worked on these three subjects.
- Improving Vehicular Network Protocols
The effectiveness of DSRC for collision avoidance depends on the communication performance of safety messages. EDCA, the standard IEEE 802.11 QoS capability, was designed for networks with a mix of voice, video and best effort traffic. We examine how to use EDCA to reduce frame collisions for a channel dominated by periodic safety messages. We have made two main contributions: Access Category (AC) Isolation and Virtual Division. AC Isolation eliminates inter-AC countdown collisions, dramatically improving the success rate for high priority packets. Virtual Division uses isolation in a novel way that also reduces collisions for lower priority packets. Both techniques are consistent with the 802.11 standard. We have done both detailed analysis and insightful NS-2 simulations.
To overcome the high packet collision probability under high traffic density as the main weakness of IEEE 802.11p, time division multiple access (TDMA) based MAC protocols have been proposed in VANET. However, considering the real two-way traffic, packet collisions still occur due to the contention or multiple vehicles using the same slot while approaching each other called encounter collisions. We proposed two TDMA based MAC protocols: MAT-MAC and PTMAC. MAT-MAC is designed for two-way traffic. It aims to reduce the number of both encounter and contention collisions while maintain high slot utilization even under unbalanced traffic scenario. PTMAC is a novel prediction based MAC protocol. Most of the encounter collisions can be predicted and potentially eliminated before they really happen. It is not only suitable for two-way traffic but also for four-way intersections in an urban area.
Delay Tolerant Networks (DTNs) are a promising solution to provide message delivery service when there are no complete end-to-end paths. Many analytic studies have been investigated to systematically understand the dynamics of DTNs and optimize the routing performance. However, most previous studies focused on replication schemes or assumed a homogeneous node mobility, which is not in line with the recent measurements of real mobility traces. We propose a stochastic control for forwarding decisions in order to minimize the expected delivery delay in a heterogeneous mobility model. We formulate the optimal forwarding problem as a Markov decision process (MDP), and then devise a greedy algorithm to obtain an optimal forwarding policy. The correctness of the algorithm is proved by induction, and the accuracy of our MDP formulation is validated by simulation using real GPS traces over 4,000 taxies in a large city. Furthermore, we compare the obtained forwarding policy with a diverse range of existing routing schemes.
- To integrate VANET with ITS infrastructure
A new two-tier BUS-VANET architecture is proposed which fully integrates traffic infrastructures with public transportation and vehicles. In this new architecture, the communications of vehicles, not only benefit from the existence of buses, but also consider the effects of using RSUs and TCC. RSUs are used to ensure service coverage while TCC is helpful for locating the destination vehicle quickly. We also investigate how much benefits can be obtained by taking advantage of traffic infrastructures. Comparing to traditional VANET, better performance can be achieved in BUS-VANET with less delivery delay and higher delivery rate.
- Develop new applications based on the future VANET and ITS infrastructure
Variety of applications have been developed taking advantage of Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communications. Unlike wired communication, wireless communication is relatively unreliable, which signiﬁcantly impacts the service quality. To further improve the transmission performance, it is necessary to investigate and deeply understand the performances of different types of applications over DSRC transmission. A real environment test-bed is developed using DSRC devices to investigate the performance of multi-hop multimedia streaming transmission. We also evaluate the performance of MapReduce application over Vehicular Cloud. Impacts of different parameters and appropriate parameter setting for Ad-Hoc Vehicle networks.
However, due to the lossy vehicular wireless links and the feature of contention based MAC protocol (CSMA/CA) employed in VANET, packet loss is a very common phenomenon. Packet collisions caused by concurrent transmissions are inevitable. This greatly impacts the video quality and makes it hard to satisfy users requirement. Moreover, since no RTS/CTS mechanism can be used for broadcasting in VANET, the lost packets cannot even be detected. Fortunately, unlike other data packets, video streaming are usually composed of a number of continuous packets with consistent sequence numbers. Therefore we can detect the lost streaming packets by checking the sequence numbers and increase the successful packet delivery rate through retransmissions. Despite of the powerful ability of reducing packet losses, aggressive retransmission may lead to additional inter-arrival delay, reduce bandwidth efficiency, block normal transmissions and even worsen the video quality. Therefore, we need to control the number of retransmissions and investigate the optimal retransmission strategies under different transmission environments. Startup caching is also necessary to ensure smooth video playing and relief the downside of retransmission.
- David H Du, PD/PI
- Sarah Sharafkandi (Ph.D. student graduated in May 2013; now works for Fitbit)
- Xiaoxiao Jiang (Ph.D. student graduated in July 2015; now working for Verizon)
- Xiang Cao, (Ph.D. student graduated in June 2017; now a faculty at Valley State University, Michigan)
- Professor Jaehoon Jeong (collaborator; a faculty at Sungkyunkwan University, Korea)
We have developed new protocols that can improve the performance of vehicular networks. We have also developed a new bus-vanet architecture that fully integrated with ITS infrastructure. We have developed a new link delay model for vehicular networks that fully considering the existence of traffic lights. We have experimented with DSRC devices and WAVE Protocol for video streaming applications. The outcomes of these research projects can improve the VANET protocol performance and to understand better how VANET can support real applications like video streaming. In addition, we have developed a self-adaptive navigation tool for cloud-based vehicular networks.
Jeong, Jaehoon, Yiwen Shen, Sangsoo Jeong, Sejun Lee, Hwanseok Jeong, Tae Oh, Taejoon Park, Muhammad Usman Ilyas, Sang Hyuk Son, and David HC Du. "STMAC: Spatio-Temporal Coordination-Based MAC Protocol for Driving Safety in Urban Vehicular Networks." IEEE Transactions on Intelligent Transportation Systems (2017).
Lee, Jinho, Jaehoon Paul Jeong, and David HC Du. "Two-way traffic link delay modeling in vehicular networks." Computer Networks 110 (2016): 253-265.
Jiang, Xiaoxiao, and David HC Du. "PTMAC: A prediction-based TDMA MAC protocol for reducing packet collisions in VANET." IEEE Transactions on Vehicular Technology 65, no. 11 (2016): 9209-9223.
Jeong, Jaehoon, Hohyeon Jeong, Eunseok Lee, Tae Oh, and David HC Du. "SAINT: self-adaptive interactive navigation tool for cloud-based vehicular traffic optimization." IEEE Transactions on Vehicular Technology 65, no. 6 (2016): 4053-4067.
Jiang, Xiaoxiao, and David HC Du. "BUS-VANET: a bus vehicular network integrated with traffic infrastructure." IEEE Intelligent Transportation Systems Magazine 7, no. 2 (2015): 47-57.
Jeong, Jaehoon Paul, Tian He, and David HC Du. "Trajectory based data forwarding schemes for vehicular networks." ZTE Communications 1 (2014): 005.
Jeong, Jaehoon Paul, Tian He, and David HC Du. "TMA: Trajectory-based Multi-Anycast forwarding for efficient multicast data delivery in vehicular networks." computer networks 57, no. 13 (2013): 2549-2563.
Jeong, Jaehoon, Shuo Guo, Yu Gu, Tian He, and David HC Du. "Trajectory-based statistical forwarding for multihop infrastructure-to-vehicle data delivery." IEEE Transactions on Mobile Computing 11, no. 10 (2012): 1523-1537.
"MA-TDMA: A Migration-based Adaptive TDMA MAC for Reducing Packet Collisions
in VANET," Proc. of VTC-Spring 2016 (with Xiaoxiao Jiang)
Lee, Jinho, Jaehoon Paul Jeong, and David HC Du. "Link Delay Modeling for Two-Way Traffic Road Segment in Vehicular Networks." In International Conference on Internet of Vehicles, pp. 339-350. Springer International Publishing, 2015. Best Paper Award
Jiang, Xiaoxiao, Xiang Cao, and David HC Du. "Multihop transmission and retransmission measurement of real-time video streaming over DSRC devices." In World of Wireless, Mobile and Multimedia Networks (WoWMoM), 2014 IEEE 15th International Symposium on a, pp. 1-9. IEEE, 2014.
Jiang, Xiaoxiao, and David HC Du. "A BUS vehicular network integrated with traffic infrastructure." In Connected Vehicles and Expo (ICCVE), 2013 International Conference on, pp. 562-567. IEEE, 2013.
Sharafkandi, Sarah, Gaurav Bansal, John B. Kenney, and David HC Du. "Using edca to improve vehicle safety messaging." In Vehicular Networking Conference (VNC), 2012 IEEE, pp. 70-77. IEEE, 2012.
Sharafkandi, Sarah, Gaurav Bansal, John Kenney, and David Du. "A novel use of EDCA to improve vehicle safety communication." In Proceedings of the ninth ACM international workshop on Vehicular inter-networking, systems, and applications, pp. 115-118. ACM, 2012.