Call for Papers
Special Issue on "Network Modeling and Simulation"
Elsevier Computer Networks Journal

Theme
=====
The rapid proliferation and integration of computer and network
systems have connected infrastructures to one another in a complex
network of interdependence.  In addition, network systems themselves
are undergoing dramatic changes in the underlying technologies and
services provided, in order to keep up with the growing demands from
new applications and users.  Scalable network modeling and simulation
approaches that can analyze and predict network behavior with known
accuracy, at varying time scales and for different network sizes and
composition, have emerged to be important (and in some cases critical)
techniques in designing, deploying, and managing such complex
networks.  Abnormal conditions such as congestion storms, network
meltdown, and behavior of badly designed protocols and policies can be
detected with analytically sound and scalable modeling and simulation
tools and corrective actions can be designed and deployed accordingly.

In the past few years, quite a number of new models of traffic,
network, and control have been created that exploit fluid flow models,
discrete event dynamical systems theory, stability theory, statistics
and measurement, and approximation theory to solve the problem of
capturing both wired and wireless network phenomena. These models,
with varying levels of abstraction, have been incorporated to develop
scalable network simulation techniques to expedite simulation of
extremely large networks (of size >= 100K nodes), rendering more than
an order of magnitude of performance improvement.  These models have
also been used to devise a wide spectrum of both passive and active
measurement techniques so as to infer, with high accuracy, various
network attributes and conditions.  Moreover, all the models and
techniques, along with an appropriate on-line network measurement
methodology, have also provided a basis for on-line network control,
dramatically reducing the time and cost required for functions such as
parameter tuning, fielding new and situation specific protocols, and
QoS provisioning. These techniques can also be harnessed to
significantly improve the efficiency of network operational functions
such as planning, service provisioning, and failure analysis.

Topics of Interest
==================
This special issue calls for articles that address the most recent
technical advances in network modeling and simulation and its use in
network measurement and control.  Of particular interests are

1. Innovative models of traffic, network, and control.
Manuscripts are sought that document innovative research on (i)
the theory and the creation of models that characterize network
traffic and/or network behavior over a broad range of network sizes
and time scales; (ii) evaluation and validation of various models with
respect to their analysis and prediction capability; and (iii)
development of methods that establish conditions under which models
are trustworthy.

Also of interest is the development of models of control that can be
deployed on top of a given network and traffic model. Control issues
of interest include, but are not limited to: network and protocol
parameter setting, distributed quality of service assignment based on
priority, dynamic selection of protocols, and distributed
implementation of control algorithms.

2. Scalable network simulation techniques.
This technical area seeks for manuscripts that report fast simulation
techniques that incorporate advances in various traffic/network models
and scaling methods to expedite network simulation of extremely large
scale.  Of particular interest are all aspects concerning
(i) the development and deployment of models that can be used to realize
faster-than-real-time simulation, while not compromising its fidelity;
and (ii) use of faster-than-real-time simulation techniques to provide
feedback to network control.

3. Model-based active and passive network measurement.
Challenges of network measurement research are to define and develop
(i) sampling techniques to support model development as well as model
validation; (ii) models that relate, with high accuracy, measurement
results with network attributes; (iii) both active and/or passive
measurement approaches that are non-intrusive and yet render
high-fidelity results; and (iv) on-line mechanisms that leverage
measurement results for network control.

Publication Schedule
====================

Manuscript due: September 15, 2004
Acceptance notification: November 30, 2004
Final manuscript due: December 31, 2004
Expected publication date: Approximately April 2005

Submission Instruction
======================

Each submission should include a cover page with the title of paper,
name and affiliation of all authors, name and contact address, fax,
and email address of the corresponding author, a list of five key
words, and a short abstract of maximum 200 words. Please send PDF
(preferred) or postscript formatted papers to Jennifer C. Hou
(jhou@cs.uiuc.edu) according to the above timetable.  All papers will
be reviewed by experts for technical merit, correctness,
relevance, and a clear exposition of English.

Submissions must meet the following criteria:

-  A paper must be material that has not been previously published
   nor is currently under review by another conference or journal.
-  Each submitted paper should be no longer than the equivalent of
   25 double space pages excluding figures, graphs, and illustrations.

Guest Editors
=============

Jennifer C. Hou
Department of Computer Science
University of Illinois at Urbana Champaign
jhou@cs.uiuc.edu

P. R. Kumar
Department of Electrical and Computer Engineering
University of Illinois at Urbana Champaign
prkumar@control.uiuc.edu