SIMULATION recent issues

Guest Editorial: Special Issue on Multi-paradigm Modeling

de Lara, J., Levendovszky, T., Mosterman, P. J. — Wed, 14 Oct 2009 03:03:38 PDT

Exploring Multi-Paradigm Modeling Techniques

Hardebolle, C., Boulanger, F. — Wed, 14 Oct 2009 03:03:38 PDT

Multi-Paradigm Modeling (MPM) addresses the necessity of using multiple modeling paradigms when designing complex systems. Because of its multidisciplinary nature, the MPM field involves research teams with technical backgrounds as different as control science, model checking, modeling language engineering or system-on-chip development. In this paper, we propose to explore the MPM domain through a survey of existing techniques from different horizons. Since the heterogeneity of models is at the heart of MPM, we first identify the sources of this heterogeneity and introduce the problems it raises. Then we show how the different existing techniques address these problems.

Patterns for Automatic Generation of Soft Real-time System Models

Florescu, O., Voeten, J., Theelen, B., Corporaal, H. — Wed, 14 Oct 2009 03:03:38 PDT

Worst-case assumptions about the timing of systems are often too conservative when analyzing distributed soft real-time systems as they lead to over-dimensioned and expensive products. For these systems, a certain percentage of deadline misses is often affordable. Instead of a binary answer regarding the schedulability of such a system, a more interesting metric is the degree to which the system meets the timing requirements. For this, an appropriate model that realistically expresses the behavior of a soft real-time system when deployed on a multiprocessor platform should be built and analyzed. In this article, we present such a modeling approach based on the formal modeling language POOSL (parallel object-oriented specification language). Moreover, to alleviate the process of modeling, we present a pattern-based description language that allows an application, together with the multiprocessor platform and the deployment to be described in a concise way. Such a pattern-based description can be translated automatically into an executable POOSL model through which performance properties can be analyzed based on simulations. The suitability of our approach is demonstrated by exploring the design space of a distributed in-car radio navigation system.

Composable Cellular Automata

Mayer, G. R., Sarjoughian, H. S. — Wed, 14 Oct 2009 03:03:38 PDT

Cellular automata (CA) provide a convenient approach to modeling a system comprised of homogeneous entities that, generally, have a spatial relationship with one another. CA are used to model systems that can be appropriately represented as a collection of interconnected automata. These networked automata may act as either a model representation of the entire system, or used to model a sub-system within a hybrid system. As the sub-systems within a hybrid system are disparate, so too can the models representing them be disparate using a multi-model approach. However, to take advantage of multi-modeling, CA and other models used to represent the sub-systems must be founded on system-theoretical principles. Furthermore, each model’s formalism must account for input and output data exchange with other modeling formalisms. Therefore, to support modular synthesis of distinct CA models with non-CA models, a composable cellular automata (CCA) formalism is proposed. This formalism is provided as a domain-neutral, mathematical specification. The CCA is then exemplified as part of a multi-model, and the GRASS development environment is used to describe one possible implementation approach.

eUDEVS: Executable UML with DEVS Theory of Modeling and Simulation

Risco-Martin, J. L., de la Cruz, J. M., Mittal, S., Zeigler, B. P. — Wed, 14 Oct 2009 03:03:38 PDT

Modeling and simulation (M&S) for system design and prototyping is practiced today both in industry and academia. M&S are two different areas altogether and have specific objectives. However, most of the time these two separate areas are taken together. The developed code is woven tightly around both the model and the underlying simulator that executes it. This constrains both the model development and the simulation engine that has an impact on the scalability of the developed code. Furthermore, a lot of time is spent in developing a model because it needs both domain knowledge and simulation techniques, which also requires communication among users and developers. The Unified Modeling Language (UML) is widely accepted in industry, whereas discrete event specification (DEVS)-based modeling that separates the model and the simulator, provides a cleaner methodology to develop models and is much used in academia. DEVS today is used by engineers who understand discrete event modeling at a highly detailed level and are able to translate requirements to DEVS modeling code. There have been earlier efforts to integrate UML and DEVS but they have not succeeded in providing a transformation mechanism owing to inherent differences in these two modeling paradigms. In this paper we present an integrated approach to cross-transformations between UML and DEVS using the proposed eUDEVS, which stands for executable UML based on DEVS. Further, we also show that the obtained DEVS models belong to a specific class of DEVS models called finite deterministic DEVS (FD-DEVS) that is available as a W3C XML schema in XFD-DEVS. We also put the proposed eUDEVS in a much larger unifying framework called the DEVS unified process that allows bifurcated model-continuity-based lifecycle methodology for systems M&S. Finally, we demonstrate the concepts with a complete example.

Formal Specification and Analysis of Domain Specific Models Using Maude

Rivera, J. E., Duran, F., Vallecillo, A. — Wed, 14 Oct 2009 03:03:38 PDT

Modeling languages play a cornerstone role in model-driven software development for representing models and metamodels. Modeling languages are usually defined in terms of their abstract and concrete syntax. This allows the rapid development of languages and some associated tools (e.g. editors), but does not allow the representation of their behavioral semantics, something especially important in certain industrial environments in which simulation and verification are critical issues. In this paper we explore the use of Maude as a formal notation for describing models, metamodels, and their dynamic behavior, making models amenable to formal analysis, reasoning, and simulation.

Engineering the Dynamic Behavior of Metamodeled Languages

Meszaros, T., Mezei, G., Charaf, H. — Wed, 14 Oct 2009 03:03:38 PDT

Language engineering is a key factor in Multi-Paradigm Modeling (MPM). Since MPM strongly builds on metamodeling, the applied language engineering methods must also be generic enough to support various metamodels. Besides the generic methods to build the abstract and concrete syntax of a visual language, only a few solutions are available to describe the dynamic behavior (‘animation’) of the models. The aim of this paper is to contribute (i) an event-based conceptual architecture to support animation, (ii) a set of visual languages to describe the animation of the models and their execution. These concepts were used to implement the animation support in our tool called Visual Modeling and Transformation System (VMTS). The VMTS animation framework introduces novel languages to describe certain aspects of animation, while integrating the benefits of the existing approaches. Our solution clearly separates the domain knowledge and the animation description both on a conceptual and implementation level. Thus, the VMTS offers a concise and systematic solution to provide a highly customizable animation framework for metamodeled languages with strong integration support to external systems such as simulation engines. The efficiency of the approach is illustrated with a rather complex animation case study implementing a model transformation debugger.

Layered Intelligence for Agent-based Crowd Simulation

Banerjee, B., Abukmail, A., Kraemer, L. — Tue, 15 Sep 2009 07:50:03 PDT

We adapt a scalable layered intelligence technique from the game industry, for agent-based crowd simulation. We extend this approach for planned movements, pursuance of assignable goals, and avoidance of dynamically introduced obstacles/threats as well as congestions, while keeping the system scalable with the number of agents. We demonstrate the various behaviors in hall-evacuation scenarios, and experimentally establish the scalability of the frame rates with increasing numbers of agents.

Towards Integration of Biological, Psychological and Social Aspects in Agent-Based Simulation of Violent Offenders

Bosse, T., Gerritsen, C., Treur, J. — Tue, 15 Sep 2009 07:50:04 PDT

In the analysis of criminal behavior, a combination of biological, psychological and social aspects may be taken into account. Dynamical modeling methods developed in recent years often address biological, psychological or social dynamical systems separately. This paper makes the first step in the development of an agent-based modeling approach for criminal behavior in which these aspects are integrated in one dynamical system. It is shown how within a certain (multi-agent) social context, biological factors, such as certain brain deviations, testosterone levels and serotonin levels, affect cognitive and emotional functioning in such a way that a crime is committed when the perceived opportunity is there. This paper presents one generic model for the behavior of violent offenders with parameters that can be set to obtain simulation traces for three known types of offenders.

VHDL-AMS Based Genetic Optimization of Mixed-Physical-Domain Systems in Automotive Applications

Wang, L., Kazmierski, T. J. — Tue, 15 Sep 2009 07:50:04 PDT

This paper presents a VHDL-AMS based genetic optimization methodology suitable for performance improvement of hardware systems in automotive applications. Models of such systems are mixed-signal (analog and digital) in which the analog parts cover mixed physical domains. A case study applying this novel method to the fuzzy logic controller (FLC) optimization in an automotive active suspension system (AASS) has been investigated. A new type of fuzzy logic membership functions with variable geometrical shapes has been proposed and optimized. In this optimization technique, VHDL-AMS is used not only for the modeling and simulation of the FLC and its underlying AASS but also for the implementation of a parallel genetic algorithm (GA). This has resulted in an integrated performance optimization system wholly implemented in the hardware description language (HDL). Results show that the proposed FLC has superior performance to that of existing FLCs that use fixed-shape membership functions.

Characterizing, Measuring, and Validating the Temporal Consistency of Live--Virtual--Constructive Environments

Hodson, D. D., Baldwin, R. O. — Tue, 15 Sep 2009 07:50:04 PDT

A distinguishing characteristic of interactive live—virtual—constructive (LVC) environments is the relaxation of data consistency to improve the performance and scalability of the underlying distributed simulation. Relaxing data consistency improves the interactive performance of the environment because autonomous distributed simulation applications can continue executing and responding to local inputs without waiting for the most current shared data values. Scalability also improves since live and simulated entities from distant geographic locations can be interconnected through relatively high-latency networks. We introduce a temporal consistency model to formally define consistency for the dynamic shared state of a LVC environment for both continuous and discrete data objects. The level of inconsistency tolerated by a LVC is found to be a function of the accuracy and timeliness requirements for the distributed data objects. These requirements are mapped to specific time intervals for which data objects are considered valid. We also develop a real-time algorithm to compute the temporal consistency of individual data objects within the LVC.

Guest Editorial: Advanced and Distributed Simulation

Liu, J. — Thu, 27 Aug 2009 08:13:46 PDT

A Hybrid HLA Time Management Algorithm Based on Both Conditional and Unconditional Information

Ke Pan, , Turner, S. J., Wentong Cai, , Zengxiang Li, — Thu, 27 Aug 2009 08:13:46 PDT

The High-Level Architecture (HLA), which is the IEEE standard for distributed simulation, defines six service groups. The Time Management (TM) service group ensures a Time-Stamp-Ordered (TSO) message delivery sequence and correct time advancement of each simulation component (federate) in a HLA-based distributed simulation application (federation). To control time advancement of a federation, a distributed TM algorithm requires each regulating federate to periodically propagate its local time information to all constrained federates for their respective calculation of the Greatest Available Logical Time (GALT). The time information propagated is called conditional information or unconditional information depending on whether it can be guaranteed to be true conditionally or unconditionally. A traditional distributed TM algorithm can be either synchronous or asynchronous. In general, a synchronous algorithm utilizes conditional information while an asynchronous algorithm utilizes unconditional information. However, both synchronous and asynchronous algorithms have their own drawbacks and thus cannot be used for all federation scenarios. To resolve the drawback of each algorithm, this paper proposes a hybrid TM algorithm by combining synchronous and asynchronous algorithms. The three algorithms have been incorporated into a Run-Time Infrastructure (RTI) and experimental results show that the hybrid algorithm effectively combines the advantages of both synchronous and asynchronous algorithms. We also compare the proposed hybrid TM algorithm with the TM algorithm implemented in the Federated Simulations Development Kit (FDK), which also uses both conditional and unconditional information. The hybrid TM algorithm is more scalable than FDK’s TM algorithm with respect to the total number of federates in a federation, because FDK’s TM algorithm has the overhead of redundant GALT calculations.

Impact of the Physical Layer Modeling on the Accuracy and Scalability of Wireless Network Simulation

Ben Hamida, E., Chelius, G., Gorce, J. M. — Thu, 27 Aug 2009 08:13:46 PDT

Recent years have witnessed a tremendous growth of research in the field of wireless systems and networking protocols. Consequently, simulation has appeared as the most convenient approach for the performance evaluation of such systems and several wireless network simulators have been proposed in recent years. However, the complexity of the wireless physical layer (PHY) induces a clear tradeoff between the accuracy and the scalability of simulators. Thereby, the accuracy of the simulation results varies drastically from one simulator to another. In this paper, we focus on this tradeoff and we investigate the impact of the physical layer modeling accuracy on both the computational cost and the confidence in simulations. We first provide a detailed discussion on physical layer issues, including the radio range, link and interference modeling, and we investigate how they have been handled in existing popular simulators. We then introduce a flexible and modular new wireless network simulator, called WSNet. Using this simulator, we analyze the influence of the PHY modeling on the performance and the accuracy of simulations. The results show that the PHY modeling, and in particular interference modeling, can have a significant impact on the behavior of the evaluated protocols at the expense of an increased computational overhead. Moreover, we show that the use of realistic propagation models can improve the simulation accuracy without inducing a severe degradation of scalability.

Models of Privacy Preserving Traffic Tunneling

Nicol, D. M., Schear, N. — Thu, 27 Aug 2009 08:13:46 PDT

Encrypted protocols, such as Secure Socket Layer (SSL), are becoming more prevalent because of the growing use of e-commerce, anonymity services, and secure authentication. Likewise, traffic analysis is becoming more common because it is often the only way to analyze these protocols. Although there are many valid uses for traffic analysis (such as network policy enforcement and intrusion detection), it can also be used to maliciously compromise the secrecy or privacy of a user. While the payload can be strongly protected by encryption, analysis of traffic patterns can yield information about the type and nature of traffic. In this paper we use simulation and analytic models to examine the impact on user experience of a scheme that masks the behavior of real traffic by embedding it in synthetic, encrypted, cover traffic. Through simulation and an analytic model we investigate the effects on the user experience using disparate and similar traffic models. This point provides a novel context where we observe the synergy of simulation and analytic modeling. We show that a detailed simulation model of network traffic characteristics can be used to estimate the parameters of an analytic model of tunneling. We see that the accuracy of the model’s predictions is directly dependent on the accuracy of parameters we obtain from the simulation. However, the simulation model does not need to have any concept of tunneling. Using simulation and analytic modeling together, we obtain an analysis whose whole is greater than the sum of its parts.

Secure Referee Selection for Fair and Responsive Peer-to-Peer Gaming

Webb, S. D., Soh, S., Trahan, J. L. — Thu, 27 Aug 2009 08:13:46 PDT

Peer-to-Peer (P2P) architectures for Massively Multiplayer Online Games (MMOG) provide better scalability than Client/Server (C/S), however, they increase the possibility of cheating. Recently proposed P2P protocols use trusted referees that simulate/validate the game to provide security equivalent to C/S. When selecting referees from un-trusted peers, selecting non-colluding referees becomes critical. Further, referees should be selected such that the range and length of delays to players is minimized (maximizing game fairness and responsiveness). In this paper we formally define the referee selection problem and propose two secure referee selection algorithms, SRS-1 and SRS-2, to solve it. Both algorithms ensure the probability of corrupt referees controlling a zone/region is below a pre-defined limit, while attempting to maximize responsiveness and fairness. The trade-off between responsiveness and fairness is adjustable for both algorithms. Simulations of three different scenarios show the effectiveness of our algorithms.

Guest Editorial: Performance Evaluation of Computer and Telecommunication Systems

Obaidat, M. S. — Thu, 13 Aug 2009 09:05:04 PDT

Using Load Transformations for the Specification of Arrival Processes in Simulation and Analysis

Heckmuller, S., Wolfinger, B. E. — Thu, 13 Aug 2009 09:05:04 PDT

In this paper the innovative approach of load transformation is used to derive valid load models for secondary loads in computer networks, as they can be observed at lower-layer interfaces of protocol stacks. We apply the technique of load transformation onto a load represented by batch Markovian arrival process (BMAP) models and choose as concrete load transformation the practically relevant token bucket mechanism, which is widely used in current networks. We are able to determine closed-form analytical solutions to describe the transformed load. Investigations by means of simulations of the quality of our analytical models reflecting token-bucket-based load transformations underline the high level of accuracy and validity of the predicted secondary load. We also present a tool implementing different transformation functionalities which can be used for the specification of arrival processes for simulative and analytic performance evaluation.

Distributed Response Time Analysis of GSPN Models with MapReduce

Haggarty, O. J., Knottenbelt, W. J., Bradley, J. T. — Thu, 13 Aug 2009 09:05:04 PDT

Generalized stochastic Petri nets (GSPNs) are widely used in the performance analysis of computer and communications systems. Response time densities and quantiles are often key outputs of such analysis. These can be extracted from a GSPN’s underlying semi-Markov process using a method based on numerical Laplace transform inversion. This method typically requires the solution of thousands of systems of complex linear equations, each of rank n, where n is the number of states in the model. For large models substantial processing power is needed and the computation must therefore be distributed. In this paper we describe the implementation of a response time analysis module for the Platform Independent Petri net Editor (PIPE2) which interfaces with Hadoop, an open-source implementation of Google’s MapReduce distributed programming environment, to provide distributed calculation of response time densities in GSPN models. The software is validated with results calculated analytically as well as simulated results for larger models. Excellent scalability is shown.

Analysis of Minimum-energy Path-preserving Graphs for Ad-hoc Wireless Networks

Rahman, A., Ahmed, M., Zerin, S. — Thu, 13 Aug 2009 09:05:04 PDT

We consider ad-hoc wireless networks and the topology control problem defined as minimizing the amount of power needed to maintain connectivity. The issue boils down to selecting the optimum transmission power level at each node based on the position information of reachable nodes. Local decisions regarding the transmission power level induce a subgraph of the maximum powered graph G_max in which edges represent direct reachability at maximum power. In this paper we propose an analysis for constructing minimum-energy path-preserving subgraphs of G_max, i.e. subgraphs minimizing the energy consumption between node pairs. We also propose an algorithm for constructing subgraph of G_max based on one-hop neighbor information. By presenting experimental results we show the effectiveness of our proposed algorithm.

Fault-tolerant Gang Scheduling in Distributed Real-time Systems Utilizing Imprecise Computations

Stavrinides, G. L., Karatza, H. D. — Thu, 13 Aug 2009 09:05:04 PDT

Distributed real-time systems play an increasingly vital role in our society. The most important aspect of such systems is the scheduling algorithm, which must guarantee that every job in the system will meet its deadline, providing high-quality (precise) results. In this paper we evaluate by simulation the performance of strategies for the scheduling of parallel jobs (gangs) in a homogeneous distributed real-time system with possible software faults. For each scheduling policy we provide an alternative version which allows imprecise computations. We propose a performance metric applicable to our problem, which takes into account the number of jobs guaranteed, as well as the precision of the results of each guaranteed job. The simulation results show that the alternative versions of the algorithms outperform their respective counterparts. To the best of our knowledge, a real-time gang scheduling approach that utilizes imprecise computations has never been discussed in the literature before.

JDMS Article Listings

Thu, 13 Aug 2009 09:05:04 PDT