Margaret L. Loper

Performance and reliability analysis of relevance filtering for scalable distributed interactive simulation

Achieving the real-time linkage among multiple, geographically-distant, local area networks that support distributed interactive simulation (DIS) requires tremendous bandwidth and communication resources. Today, meeting the bandwidth and communication requirements of DIS is one of the major challenges facing the design and implementation of large scale DIS training exercises. In this article, we discuss the DIS scalability problem, briefly overview the major bandwidth reduction techniques currently being investigated and implemented in contemporary DIS systems, and present a detailed analysis on the performance and reliability of relevance filtering—a promising technique to improve the scalability of distributed simulation. The idea of relevance filtering is to analyze the semantic contents of the state update messages of a simulated entity (vehicle) and transmit only the ones found to be relevant to other entities. We present our entity-based model for relevance filtering and discuss the implementation of filtering-at-transmission and filtering-at-reception. We introduce the concept of filtering reliability and present different methods to eliminate or reduce filtering errors. Methods that can ensure complete filtering reliability while providing significant bandwidth reduction are developed. Performance evaluation results of relevance filtering and of the filtering reliability methods are presented. The insight gained from our work and the challenges still facing the design of large scale DIS training exercises are discussed.

Multi-level Modeling of Complex Socio-Technical Systems

Abstract The application of multi-level modeling approaches to enterprise systems modeling can support leadership decision making for transformation of defense department operations. In particular, this paper presents a framework for modeling enterprise transformation applied to complex analysis of defense in an era of technology globalization. Large-scale public-private enterprises can generally be described as complex adaptive systems. Multi-level modeling of complex adaptive systems is considered as an approach to inform leadership of actions and consequences of decisions addressing transformative change. An example is elaborated for the domain of secure defense communications. A framework for modeling the secure communications enterprise presents an interesting case for exploration of both multi-level modeling approaches and potential policy guidance at a global level. Finally, the use of these types of models to enable policy exploration is illustrated for the particular case of counterfeit parts in the supply chain of military communication systems. The paper concludes with recommendations for further work in validating the modeling approach and application of the approach to a sample policy flight simulator representing counterfeit parts policy in military systems.

Pre-sampling as an approach for exploiting temporal uncertainty

We describe an approach to exploit temporal uncertainty in parallel and distributed simulation by utilizing time intervals rather than precise time stamps. Unlike previously published work that proposes new message ordering semantics, our approach is based on conservative, time stamp order execution and enhancing the lookahead of the simulation by pre-drawing random numbers from a distribution that models temporal uncertainty. The advantages of this approach are that it allows time intervals to be exploited using a conventional Time Stamp Order (TSO) delivery mechanism, and it offers the modeler greater statistical control over the assigned time stamps. An implementation of this approach is described and initial performance measurements are presented.

Grand challenges in modeling and simulation: expanding our horizons

There continues to be many advances in the theory and practice of Modeling and Simulation (M&S). However, some of these can be considered as Grand Challenges; issues whose solutions require significant focused effort across a community, sometimes with ground-breaking collaborations with new disciplines. In 2002, the first M&S Grand Challenges Workshop was held in Dagstuhl, Germany, in an attempt to focus efforts on key areas. In 2012, a new initiative was launched to continue these Grand Challenge efforts. Panel members of this third Grand Challenge present their views on M&S Grand Challenges. Themes presented in this panel include M&S Methodology; Agent-based M&S; M&S in Systems Engineering; Cyber Systems Modeling; and Network Simulation.

A case study in exploiting temporal uncertainty in parallel simulations

Approximate Time (AT) has been proposed as a means for expressing temporal uncertainty in distributed simulation applications in order to enhance parallel performance. This is accomplished by specifying time intervals rather than precise values to indicate when an event might occur. This paper describes a case study in applying AT to a queueing simulation application in order to assess the performance and accuracy that can be obtained using this technique. Up to an order of magnitude speedup was obtained, with error in throughput statistics less than 3%, although somewhat larger error was reported in delay statistics.

Synchronizing simulations in distributed interactive simulation

Distributed interactive simulation (DIS) provides the infrastructure to build large-scale simulations by interconnecting independent simulators via a network. These simulators execute primarily in real time because of the human-in-the-loop requirements. The future of DIS in the general area of distributed simulation will include non real time simulations, such as constructive wargames. Active research in the area of parallel discrete event simulation (PDES) has devised means by which event driven simulations can be parallelized, by utilizing state of the art parallel architectures and scheduling events which can be executed concurrently and in a conflict-free fashion. Simulation has experienced growth in both these communities. Each community has specific needs and requirements which have to be met. This paper examines the areas of DIS in which PDES techniques may be employed to achieve more parallelism. In particular, this paper addresses the need for, and implementation of synchronization in current DIS networks. The interaction between constructive and other categories of simulation will require synchronized DIS applications, and employ mechanisms utilized in PDES.

Applying temporal databases to HLA data collection and analysis

The High Level Architecture (HLA) for distributed simulations was proposed by the Defense Modeling and Simulation Office of the Department of Defense (DOD) in order to support interoperability among simulations as well as reuse of simulation models. One aspect of reusability is the ability to collect and analyze data from simulation executions, including a record of events that occur during the execution, and the states of the simulation objects. Several approaches have been developed for data collection in distributed simulation environments. The HLA presents an interesting new paradigm within which to design effective data collection and analysis techniques. The capabilities of the run-time infrastructure (RTI) can be exploited to design efficient and flexible data collection tools. Recent research on the efficient log-based implementations of temporal databases may enable more efficient collection and analysis of data from simulation executions. Using a distributed real-time temporal database approach, we may be able to expand run-time analysis opportunities. We present a list of important research questions regarding the utility and viability of a temporal database data collection and analysis approach. The questions address issues of how and where to collect the data, how to resolve temporal issues in a distributed system, what functionality must be supported by the temporal database, and what the relationship to the run-time infrastructure (RTI) should be.

Modeling an emergency operations center with agents

The Georgia Emergency Management Agency (GEMA) operates the State Operations Center, which is the centralized hub for all communications regarding state operations, with respect to emergency response. The center provides three primary functions: request for information, request for assistance, and media relations. The center has responsibility for the entire process of servicing an emergency, from instigation to communication to execution to ending service for the emergency. To accomplish this, the center is set up like a production facility, which follows a well-defined process articulated in a plan. In this project, we worked with GEMA to develop a simulation of the center. The simulation, called GEMASim, models the information flow among people working within the center, as well as the flow of information coming into and leaving the center. Data is presented on the initial experiments with GEMASim using two different scenarios.

A framework for simulating human cognitive behavior and movement when predicting impacts of catastrophic events

Our nation has seen an increased need to train its civil authorities and emergency personnel under life-threatening scenarios where human life and critical infrastructure are assumed to be at risk. This training is typically obtained or re-enforced via (human) performance-based tests. At issue is the ability to accurately simulate the scenarios without exposing personnel or human test subjects to injury. In addition, these performance-based tests carry a large monetary cost, and certain scenarios are so complicated, catastrophic or rare that any performance-based test is unrealistic. Our paper outlines the research that must be conducted to develop a framework for modeling and analyzing risk-assessment and decision making when evacuating large populations. The research is aimed at extending an existing construct for simulating passenger and crew behavior during aircraft evacuations, to larger populations, and relies upon rare-event simulation methods, parallel-and-distributed simulation and agent-based simulation.

Exploiting temporal uncertainty in process-oriented distributed simulations

Existing research has defined a new type of simulation time called Approximate Time, where the simulations knowledge about the values that represent time is uncertain. The approach is based on temporal uncertainty and uses time intervals rather than precise time values to represent time. Simulation language constructs are necessary to provide a convenient means of exploiting the temporal uncertainty to simulation modelers. To address this problem, a new time advance primitive for process-oriented simulations was developed, termed the interval hold construct. Interval hold is an extension of the well-known hold primitive used in conventional simulation languages. This paper defines the interval time advance primitive and describes an algorithm for implementing it.