Proximity Detection in Distributed Simulation of Wireless Mobile Systems Luciano Bononi Michele Bracuto Gabriele D’Angelo Lorenzo Donatiello Dipartimento di Scienze dell’Informazione, Università degli Studi di Bologna Mura Anteo Zamboni 7, 40127, Bologna, Italy {bononi, bracuto, gdangelo, donat}@cs.unibo.it ABSTRACT The distributed and the Grid Computing architectures for the simulation of massively populated wireless systems have recently been considered of interest, mainly for cost reasons. Solutions for generalized proximity detection for mobile objects is a relevant problem, with a big impact on the design and the implementation of parallel and distributed simulations of wireless mobile systems. In this paper, a set of solutions based on tailored data structures, new techniques and enhancements of the existing algorithms for generalized proximity detection are proposed and analyzed, to increase the efficiency of distributed simulations. The paper includes the analysis of computation complexity of the proposed solutions and the performance evaluation of a testbed distributed simulation of ad hoc network models. Recent works have shown that the performance of distributed simulation of dynamic complex systems could benefit from a runtime migration mechanism of model entities, which reduces the communication overheads. Such migration mechanisms may interfere with the generalized proximity detection implementations. The analysis performed in this paper illustrates the effects of many possible compositions of the proposed solutions, in a real testbed simulation framework. Categories and Subject Descriptors I.6.8 [Simulation and Modeling]: Types of Simulation – discrete event, distributed. General Terms Algorithms, Performance, Design, Experimentation. Keywords Proximity detection, wireless systems, distributed simulation, data distribution management. 1. INTRODUCTION The computer simulation is a powerful technique to model, evaluate and predict the behavior of complex dynamic systems. Many systems are composed by a great number of entities, dynamically interacting in unpredictable way. The analysis approach based on mathematical modeling is often inadequate to express the fine details and complex interactions of such systems. The simulation can be considered a viable solution in these cases. A discrete event simulation mimics the behavior of a system model at discrete points in time. The interactions between simulated entities are represented by simulation events and the processing of such events causes the evolution of the simulation process [9]. A common problem of many simulations is to determine what entities are involved in the notification and processing of a given event. As an example, in a wireless simulation a new frame transmission causes the creation of one transmission event- message by the transmitter entity, that must be delivered in chronological (causal) order to the whole set of potential receiver entities. Due to the local broadcast nature of the wireless transmission, wireless signals decay with distance depending on propagation model assumptions [14]. For this reason, it makes sense to distribute transmission event messages only to the subset of neighbor hosts that will be reached by the transmission effects [12, 13]. Since hosts may be mobile, such subset is dynamic. A generalization of the subset definition problem is known in the literature as generalized proximity detection for moving objects [15]. Under the simulation viewpoint, the problem can be translated in the dynamic identification of the recipient entities of each event message. In a distributed simulation, this operation is usually executed by the event distribution management component. The conflict detection between moving objects is a more specific case of the general proximity detection algorithm [17]. A partially related field is given by physical particles’ simulation: to determine the evolution of a single simulated entity it is necessary to take into account its interactions (forces) with all other entities in the system [1]. In the following, we will focus on distributed simulation of wireless models. Some work has been done in the past to address the proximity detection problem over single processor architectures [13], and multiprocessor architectures with shared memory [10,12,16]. On the other hand, distributed and Grid Computing architectures are gaining a lot of interest, mainly for Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. MSWiM'06, October 2–6, 2006, Torremolinos, Malaga, Spain. Copyright 2006 ACM 1-59593-477-4/06/0010...$5.00. 44