Institute for Theoretical Information Technology

Stochastic Networks - Analysis and Evaluation Supported by Modern Simulation Tools

Dates

Lab Dates

Prerequisites

• Sufficient experience in object-oriented programming is highly recommendable. C++ will be used throughout the lab, e.g., you should be able to work with classes, typecasting, pointers and references, etc.
• Basic knowledge of probability theory is desirable.
• Knowledge of queuing systems and stochastic networks is advantageous, but it will be acquainted in the course of the lab.

Content

Within this laboratory you will learn to utilize the modern event-driven network simulator OMNeT++ to simulate and evaluate stochastic networks. Diverse approaches to the design and evaluation of stochastic networks will be worked out in the course of the lab. A particular focus of this lab is to compare theoretical and practical results to understand the advantages and limitations of mathematical modelling and simulation, respectively.

The theoretical part will cover:

• Statistical methods to determine the reliability of the simulated results
• Markov Chains, Markov processes, particularly Poisson and birth-death processes
• Queuing Systems
• Queuing Networks, particularly Jackson and BCMP networks

Typical practical problems are:

• Reliability and survival analysis in multi-component systems
• Gilbert Elliott's satellite channel
• ALOHA, Slotted ALOHA and CSMA
• Tandem systems
• Router: IP address distribution
• Radiation detection
• Google page rank
• Queues in supermarkets or cafeterias
• Billing monitoring
• Traffic simulation for vehicle to vehicle (V2V) communications
• Smart grid communications

Material

Corresponding literature will be announced at the beginning of the lab.

• Lecture notes (German): Theoretische Informationstechnik
• Lecture notes (German): Kommunikationsnetze II
• Lecture notes (German): Stochastik für Informatiker

During the lab we will provide handouts on the script, appendix as well as homework and lab-tasks.

• Lab1: A simple Full-Adder
• Lab2: Probabilistic Analysis of System Reliability
• Lab3: Probabilistic Analysis of System Survival
• Lab4: Markov Chains and Elementary Queuing Systems
• Lab5: Queueing Systems/Networks - MENSA Field Study
• Lab6: Vehicle-to-Vehicle Communications

Examination

Mandatory for passing the laboratory course:

• Solving problems throughout the course.
• Passing tests on the theoretical background.

Contact

Dr.-Ing. Michael Reyer upon agreement
M.Sc. Jose Angel Leon Calvo upon agreement
Dipl.-Ing. Christopher Schnelling, M.Sc. upon agreement

zurück zur Liste der Vorlesungen