Home | Publications | Current Projects Funding Students Past Projects

Currently Active Projects
Vehicular Networking | Network Simulation | Traffic Generation | High-Speed TCP | IEEE 802.11e

Vehicular Networking

ODU's VANET Research Group | ODU's Transportation Research Cluster

S. Rizvi, S. Olariu, M.C. Weigle, and M. Rizvi, A Novel Approach to Reduce Traffic Chaos in Emergency and Evacuation Scenarios, Proceedings of the IEEE Vehicular Technology Conference - Fall, Baltimore, MD, October 2007.

This paper proposes a novel chaos reducing information dissemination approach for spatio-temporal traffic information related to first responders and evacuation scenarios using Vehicular Ad Hoc Networks (VANETs). In our approach, we provide an emergency vehicle path clearing technique. Therefore, traffic confusion and chaos is lowered on evacuation and emergency vehicle routes. Simulation results show that our approach works efficiently without fully relying on any message relaying infrastructure.

K. Ibrahim and M.C. Weigle, Accurate Data Aggregation for VANETs (poster), Proceedings of ACM VANET, Montreal, Canada, September 2007, pp. 71-72.

Data aggregation is an important issue for vehicular ad-hoc networks (VANETs). Congestion notification applications are built to warn drivers of traffic slowdowns far enough in advance that the drivers may take alternate routes. Data that is broadcast should be self-contained and fit into a single MAC-layer frame. With dense traffic, aggregation is needed to represent a large number of vehicles in relatively small frame. We present a new technique for aggregating vehicles' data without losing accuracy. Vehicles build a local view based on speed and position reports from neighboring ve- hicles. This local view, representing vehicles up to 1.6 km ahead, is then aggregated into a single frame and broad- cast. Vehicles use received aggregated frames to extend their views even farther.

G. Yan, G. Choudhary, M.C. Weigle, and S. Olariu, Providing VANET Security Through Active Position Detection (poster), Proceedings of ACM VANET, Montreal, Canada, September 2007, pp. 73-74.

Our main contribution is a novel approach to enhancing position security in VANET. We achieve local and global position security by using the on-board radar to detect neighboring vehicles and to confirm their announced coordinates. We compute cosine similarity among data collected by radar and neighbors' reports to filter the forged data from the truthful data. Based on filtered data, we create a history of vehicle movement. By checking the history and computing similarity, we can prevent a large number of Sybil attacks and some combinations of Sybil and position-based attacks.

M.C. Weigle and S. Olariu, Intelligent Highway Infrastructure for Planned Evacuations, Proceedings of the First International Workshop on Research Challenges in Next Generation Networks for First Responders and Critical Infrastructures (NetCri), New Orleans, LA, April 2007, pp. 594-599.

Disasters, natural and man-made alike, pose a serious threat to the nation by taking a heavy toll in human lives, destroying the public infrastructure and production capacity, interrupting supply lines, and stalling economic activity. One of the time-honored strategies for dealing with predictable natural disasters is a planned evacuation of the population from the afflicted area. Thus, evacuation strategies and supporting infrastructure are of the highest importance for mitigating the effects of such events. The main contribution of this work is to propose an intelligent highway infrastructure in support of planned evacuations. Specifically, we show that the recently-proposed Architecture for the Notification of Traffic Incidents and Congestion (NOTICE) can be enhanced to support the needs of large-scale evacuations.

Network Simulation (more info)

Recent contributions to the ns-2 network simulator: Full-TCP and Wireless, DelayBox, and PackMime-HTTP

M.C. Weigle, Improving Confidence in Network Simulations (invited paper), Proceedings of the Winter Simulation Conference, Monterey, CA, December 2006, pp. 2188-2194.

Simulation has become the tool of choice for an increasing number of networking researchers. Unfortunately, standard statistical techniques often cannot be applied with Internet-like heavy-tailed workloads are used as input. We present issues involved in using heavy-tailed distributions in network simulations, including three different methods for dealing with such distributions in simulation. We also discuss the proper use of the random number generator implemented in the ns-2 simulator and the impacts of improper usage.

Synthetic Internet Traffic Generation

M.C. Weigle, P. Adurthi, F. Hernandez-Campos, K. Jeffay, and F.D. Smith, Tmix: A Tool for Generating Realistic Application Workloads in ns-2, ACM Computer Communication Review, July 2006, Vol 36, No 3, pp. 67-76.

The tmix Internet traffic model, developed in the Department of Computer Science at UNC, is based on actual Internet traffic traces. The traces are analyzed and a model (in the form of a set of connection vectors) is produced. tmix is currently used in the UNC network testbed for evaluating network protocols. Our work involves porting the tmix network testbed traffic generator for use in the popular ns-2 network simulator. Once we have validated our implementation in ns-2, we plan to release the source code. Researchers using this tool could generate traffic in their simulations representative of traffic from a particular network trace. In the future, we plan to host a repository of connection vectors from traces on various links in the Internet.

High-Speed TCP Protocols (more info)

MS Projects (Fall 2007)

M.C. Weigle, P. Sharma, and J. Freeman, Performance of Competing High-Speed TCP Flows, Proceedings of NETWORKING, Coimbra, Portugal, May 2006, pp. 476-487.

Recently, several new implementations of TCP have been developed with the goal of allowing scientists to take advantage of high capacity networks in order to share large datasets efficiently. It has been shown that standard TCP, which handles most Internet traffic, has problems when attempting to send data at very high speeds (e.g., 10 Gbps). These new high-speed variants of TCP were designed to solve the problems with high-speed transfers while maintaining TCP's reliability and method of congestion control. The most prominent of these are High-Speed TCP, Scalable TCP, FAST TCP, and BIC-TCP. Previous work has looked at these high-speed TCP implementations and how they perform individually on dedicated links. More recent work has focused on how these implementations fairly share link bandwidth with traditional TCP. We are interested in investigating how these protocols compete with each other and with traffic from standard Internet applications, such as web browsers. Additionally, we would like to investigate how Active Queue Management (AQM) techniques affect the performance of these protocols. AQM has been shown to smooth loss patterns, reducing the incidence of consecutive lost packets from a single flow.

Evaluation of IEEE 802.11e Wireless LANs (more info)

M. Thottan and M.C. Weigle, Impact of 802.11e EDCA on Mixed TCP-based Applications, Proceedings of the International Wireless Internet Conference (WICON), Boston, MA, August 2006.

There has been an explosive growth in the use of wireless LANs (WLANs) to support network applications ranging from web-browsing and file-sharing to voice calls. It is difficult to optimally configure WLAN components, such as access points (APs), to meet the quality-of-service requirements of the different applications, as well as ensuring flow-level fairness. Recent work has shown that the widely-deployed IEEE 802.11 MAC Distributed Coordination Function (DCF) is biased against downstream flows. The new IEEE 802.11e standard introduces QoS mechanisms, such as Enhanced Distributed Channel Access (EDCA), that allow this unfairness to be addressed. So far, only limited work has been done to evaluate the impact of these MAC protocols on TCP-based applications. In this paper, through ns-2 simulations, we evaluate the impact of EDCA on TCP application traffic consisting of both long and short-lived TCP flows. We find that the performance of TCP applications is very dependent upon the settings of the EDCA parameters and buffer lengths at the AP. We also show that the performance of the admission control strategy employed depends on the buffer lengths at the AP and the traffic intensity.

Return to top

Funding

Return to top

Current Students

 PhD Students
  • Khaled Ibrahim
  • Gongjun Yan - co-advised with Stephan Olariu
  • Martin Klein - committee member (advisor: Michael Nelson)
  • Filip Cuckov (ECE) - committee member (advisor: Min Song)
     
    MS Students
  • Aditya Varakantam (MS Project, expected Aug 2008)
  • Raghevendra Aekka (MS Project, expected Dec 2008)
     
    Undergrads
  • Matt Crinklaw-Vogt
  • Colin Courtney

    • Former Students

  • Sumanth Gelle (MS Project: Performance of High-Speed TCP Protocols over NS-2 TCP Linux, May 2008) - AH Computer Consulting
  • Omkar Kasinadhuni (MS Project: TCP Validation, May 2008) - Infosys, Seattle, WA
  • Siva Tella (MS Project: Performance of Competing High Speed TCP Flows with Background Traffic, May 2008) - AH Computer Consulting
  • Gyanesh Choudhary (MS Project: Providing VANET Security Through Active Position Detection, Dec 2007) - GE, Charlottesville, VA
  • Hadi Arbabi (MS Thesis: Channel Management in Heterogeneous Cellular Networks, Aug 2007) - committee member
  • Prashanth Adurthi (MS Thesis: Generating Tmix-Based TCP Application Workloads in ns-2 and GTNetS, Aug 2006) - EMC, Raleigh, NC
  • Pankaj Sharma (MS Thesis: Performance Analysis of High-Speed Transport Control Protocols, Aug 2006)

    • Return to top

    Past Projects

    Performance of Active Queue Management Techiniques over Multiple-Congested Links (more info)

    M.C. Weigle, D. Vembar, and Z. Du, Assessing the Impact of Multiple Active Queue Management Routers, Proceedings of Globecom, San Francisco, CA, November 2006.

    Recent studies have shown that a non-negligible number of packets face multiple congested links on Internet paths. We investigate the impact of using multiple Active Queue Management (AQM) routers on paths that consist of multiple congested links. We present the results of an ns-2 evaluation study of various AQM techniques that, in contrast to previous studies of AQM, uses a complex network topology including up to 3 congested links, reverse path traffic, and realistic round-trip times. We consider the effects of multiple AQM routers on the throughput of long-lived FTP flows and HTTP response times of web traffic. We find that, especially for web traffic, performance is improved as the number of AQM routers is increased, but that significant improvements occur with even a single AQM router.

    Network Simulation

    TCP Early Congestion Detection and Reaction

    M.C. Weigle, K. Jeffay, and F.D. Smith, Delay-Based Early Congestion Detection and Adaptation: Impact on web performance, Computer Communications, May 2005, Vol 28/8, pp. 837-850.

    Concerns over the scalability of TCP's end-to-end approach to congestion control and its AIMD congestion adaptation have led to proposals for router-based congestion control, specifically, active queue management (AQM). In this paper we present an end-to-end alternative to AQM - a new congestion detection and reaction mechanism for TCP based on measurements of one-way transit times of TCP segments within a TCP connection. Our design, called Sync-TCP, places timestamps in TCP headers, measures variation in one-way transit times, and uses these measurements as a form of early congestion notification. We demonstrate empirically that: (1) Sync-TCP provides better throughput and HTTP response-time performance than TCP Reno, (2) Sync-TCP provides better early congestion detection and reaction than the Adaptive Random Early Detection with Explicit Congestion Notification AQM mechanism, (3) Sync-TCP's congestion detection and adaptation mechanisms are robust against clock drift, (4) Sync-TCP is an incrementally deployable protocol - Sync-TCP connections can co-exist with TCP Reno connections in a network, and (5) the performance of TCP Reno connections are improved with the addition of even a small percentage of Sync-TCP connections.

    M.C. Weigle, Investigating the Use of Synchronized Clocks in TCP Congestion Control, Ph.D. Dissertation, Technical Report TR03-038, Department of Computer Science, University of North Carolina at Chapel Hill, August 2003.

    This work provides a method for measuring a flow's one-way transit time (OTT) and returning this exact timing information to the sender, a comparison of several methods for using OTTs to detect congestion, a family of end-to-end congestion control mechanisms (called Sync-TCP) based on using OTTs for congestion detection, and study of standards-track TCP congestion control and error recovery mechanisms in the context of HTTP traffic.

    Return to top

    QualNet Network Simulator University Program QualNet Network Simulator