Showing papers by "Richard M. Fujimoto published in 2005"

Efficiency of Simulated Vehicle-to-Vehicle Message Propagation in Atlanta, Georgia, I-75 Corridor

Abstract: Exploitation of in-vehicle information technology (e.g., mobile computing and wireless communications) in surface transportation systems is a clearly emerging trend. Equipping vehicles with computing, communication, and sensing capabilities presents significant opportunities for a vast array of transportation services. Vehicle-to-vehicle (V2V) communication may be considered for applications such as incident detection, crash reporting, traveler information dissemination, and network operations. In-vehicle computing systems facilitate the customization of information services to the needs and characteristics of individual travelers. In addition, these systems allow coverage to extend beyond areas where roadside equipment has been placed. This study provides the initial investigation needed to test the feasibility of these advanced communication networks. Several observations may be drawn from the study. First, V2V communication is a feasible way to propagate information along the I-75 freeway in the Atlant...

An empirical study of short range communications for vehicles

Abstract: This paper presents a detailed measurement study of short range communications between vehicles and between vehicles and roadside stations in a realistic highway scenario. We show the expected wireless communication characteristics in a driving environment and identify factors that significantly affect communication performance. We also illustrate the benefits of multi-hop communication in improving communication performance.

Parallel Event-Driven Neural Network Simulations Using the Hodgkin-Huxley Neuron Model

Abstract: Neural systems are composed of a large number of highly-connected neurons and are widely simulated within the neurological community. In this paper, we examine the application of parallel discrete event simulation techniques to networks of a complex model called the Hodgkin-Huxley neuron[1]. We describe the conversion of this model into an event-driven simulation, a technique that offers the potential of much greater performance in parallel and distributed simulations compared to time-stepped techniques. We report results of an initial set of experiments conducted to determine the feasibility of this parallel event-driven Hodgkin-Huxley model and analyze its viability for large-scale neural simulations.

Optimistic Parallel Discrete Event Simulations of Physical Systems Using Reverse Computation

Abstract: Efficient computer simulation of complex physical phenomena has long been challenging due to their multi-physics and multi-scale nature. In contrast to traditional time-stepped execution methods, we describe an approach using optimistic parallel discrete event simulation (PDES) and reverse computation techniques. We show that reverse computation-based optimistic parallel execution can significantly reduce the execution time of a plasma simulation without requiring a significant amount of additional memory compared to conservative execution techniques. We describe an application-level reverse computation technique that is efficient and suitable for complex scientific simulations involving floating point operations.

Analysis and design of vehicular networks

Abstract: Advances in computing and wireless communication technologies have increased interest in "smart" vehicles---vehicles equipped with significant computing, communication and sensing capabilities to provide services to travelers. Smart vehicles can be exploited to improve driving safety and comfort as well as optimize surface transportation systems. Wireless communications among vehicles and between vehicles and roadside infrastructures represent an important class of vehicle communications. One can envision creating an integrated radio network leveraging various wireless technologies that work together in a seamless fashion. Based on cost-performance tradeoffs, different network configurations may be appropriate for different environments. An understanding of the properties of different vehicular network architectures is absolutely necessary before services can be successfully deployed. Based on this understanding, efficient data services (e.g., data dissemination services) can be designed to accommodate application requirements. This thesis examines several research topics concerning both the evaluation and design of vehicular networks. We explore the properties of vehicle-to-vehicle (v2v) communications. We study the spatial propagation of information along the road using v2v communications. Our analysis identifies the vehicle traffic characteristics that significantly affect information propagation. We also evaluate the feasibility of propagating information along a highway. Several design alternatives exist to build infrastructure-based vehicular networks. Their characteristics have been evaluated in a realistic vehicular environment. Based on these evaluations, we have developed some insights into the design of future broadband vehicular networks capable of adapting to varying vehicle traffic conditions. Based on the above analysis, opportunistic forwarding that exploit vehicle mobility to overcome vehicular network partitioning appears to be a viable approach for data dissemination using v2v communications for applications that can tolerate some data loss and delay. We introduce a methodology to design enhanced opportunistic forwarding algorithms. Practical algorithms derived from this methodology have exhibited different performance/overhead tradeoffs. An in-depth understanding of wireless communication performance in a vehicular environment is necessary to provide the groundwork for realizing reliable mobile communication services. We have conducted an extensive set of field experiments to uncover the performance of short-range communications between vehicles and between vehicles and roadside stations in a specific highway scenario.

Simulated Vehicle-to-Vehicle Message Propagation Efficiency on Atlanta's I-75 Corridor

Abstract: Exploitation of in-vehicle information technology (e.g., mobile computing and wireless communications) in surface transportation systems is a clear emerging trend. Equipping vehicles with computing, communication, and sensing capabilities presents significant opportunities for a vast array of transportation services. Vehicle-to-vehicle communication may be considered for applications such as incident detection, traveler information dissemination, network operations, etc. In-vehicle computing systems facilitate the customization of information services to the needs and characteristics of individual travelers. In addition, in-vehicle systems allow coverage to extend beyond areas where roadside equipment has been placed. This study provides the initial investigation needed to test the feasibility of these advanced communication networks. Several observations may be drawn from this study. First, vehicle-to-vehicle communication is a feasible way to propagate information along freeways in metro areas, although propagation performance depends critically on factors such as the density of instrumented vehicles along the end-to-end path. Second, the simulation methodology described within this study allows one to estimate the minimum required fleet penetration ratio for effective communication, given the traffic density and application requirements. Third, the message propagation delay is highly variable when instrumented vehicle density is low. A particular delay may be well below or above the average depending on traffic conditions. For applications requiring highly reliable, minimal message propagation it may be necessary to design networks that provide extra support to avoid such variation. Future research is required to examine additional traffic conditions (e.g., congestion due to an incident) and study the effectiveness of this approach for particular applications. TRB 2005 Annual Meeting CD-ROM Paper revised from original submittal. Wu, H., J. Lee, M. Hunter, R. Fujimoto, R. Guensler, J. Ko 3

Performance prediction of large-scale parallel discrete event models of physical systems

Abstract: A virtualization system is presented that is designed to help predict the performance of parallel/distributed discrete event simulations on massively parallel (supercomputing) platforms. It is intended to be useful in experimenting with and understanding the effects of execution parameters, such as different load balancing schemes and mixtures of model fidelity. A case study of the virtualization system is presented in the context of plasma physics simulations, highlighting important virtualization challenges and issues, such as reentrancy and synchronization in the virtual plane, and our corresponding solution approaches. A trace-based prediction methodology is presented, and is evaluated with a 1-D hybrid collisionless shock model simulation, with the predicted performance being validated against one obtained in actual simulation. Predicted performance measurements show excellent agreement with actual performance measurements on parallel platforms containing up to 512 CPUs.

Vehicular networks in urban transportation systems

Abstract: Recent results concerning the development and exploitation of wireless vehicle-to-vehicle and vehicle-to-infrastructure communications in future generation intelligent transportation systems are discussed, as well as simulation techniques and tools used to effectively model such systems. Specifically, new algorithms to disseminate information in vehicle-to-vehicle networks have been developed, and their performance evaluated using simulation modeling of traffic along the I-75 corridor in the Atlanta metropolitan area. Work to validate the simulations models is summarized. Field experiments with wireless communication along I-75 are described, as well as extensions of these results to investigate communication architectures for vehicular networks. Results concerning multi-resolution simulation of urban transportation networks involving the use of aggregated data from regional planning models in microscopic traffic simulations are summarized.

An architecture study of infrastructure-based vehicular networks

Abstract: There are various ways to organize and connect in-vehicle computing systems by exploiting existing wireless technologies. It is desired to provide a wireless infrastructure for vehicles (e.g., offering reliable broadband channels, transportation related services and Internet access). In this paper we identify several design options for infrastructure-based vehicular networks and evaluate these designs in a realistic vehicular environment.

A New Simulation Technique for Study of Collisionless Shocks: Self‐Adaptive Simulations

Abstract: The traditional technique for simulating physical systems modeled by partial differential equations is by means of time‐stepping methodology where the state of the system is updated at regular discrete time intervals. This method has inherent inefficiencies. In contrast to this methodology, we have developed a new asynchronous type of simulation based on a discrete‐event‐driven (as opposed to time‐driven) approach, where the simulation state is updated on a “need‐to‐be‐done‐only” basis. Here we report on this new technique, show an example of particle acceleration in a fast magnetosonic shockwave, and briefly discuss additional issues that we are addressing concerning algorithm development and parallel execution.

Using Ghosts for Global Topology Knowledge in Space-Parallel Distributed Network Simulations

Abstract: The authors discuss an approach to federated network simulations that eases the burdens on the simulation developer in creating space-parallel simulations. Previous approaches have had difficulties that arise from the need for global topology knowledge when forwarding simulated packets between federates. In all but the simplest cases, proper packet-forwarding decisions between federates requires routing tables of size O(mn), where m is the number of nodes modeled in a particular federate, and n is the total number of network nodes in the entire topology. Furthermore, the benefits of the well-known NIx-Vector routing approach cannot be fully achieved without global knowledge of the overall topology. The authors overcome these difficulties by using a topology partitioning methodology that uses ghost nodes. They show experimentally that the memory overhead associated with ghosts is minimal relative to the overall memory footprint of the simulation.