Analysis and Design of Wireless Networks: A Stochastic Geometry Approach

This course gives an in-depth and self-contained introduction to stochastic geometry and random graphs, applied to the analysis and design of modern wireless systems. In many such systems, including cellular, ad hoc, sensor, and cognitive networks, users or terminals are mobile or deployed in irregular patterns, which introduces considerable uncertainty in their locations. Since signal powers and interference depend critically on the node distances, it is essential that the network geometry be accurately modeled for analysis and design purposes. As a consequence, stochastic geometry and the theory of random geometric graphs have emerged as essential tools for research in wireless networks. In the last decade, these techniques have led to important results and insights on the connectivity, capacity, and fundamental limits of wireless networks, and the coverage of sensor networks, and, in view of new emerging types of wireless networks such as cognitive and mesh networks and femtocell and relaying approaches to enhance cellular systems, their importance will only increase.

SPEAKER
Prof. Martin Haenggi, University of Notre Dame

REGISTRATIONS AND PRICE
Please, register through the link on this page. Participants are taken in the registration order, as long as there is free space in the class room.
This course is free-of-charge for INFORTE member organisation's staff. For others, participation fee is 360 EUR. You can check the list of the member organizations when filling in the registration form.

LOCATION AND ACCOMMODATION POSSIBILITIES
This event will be organized in Oulu, in Lasaretti hotel, room Rossi. Lasaretti is within 10-15 minutes walking distance from the city centre. Lasaretti has free-of-charge parking space. If needed, you can ask accommodation from Lasaretti (Hansel price is 93,40e/night/single room - for others 125e/night). There is a quata booked for our participants - the quata expires on 24th of August.

SCHEDULE

Thursday
10:00 Coffee & Opening
10:15 Introduction and a Key Result
11:45 Lunch break
12:45 Throughput Analysis and Design Aspects
14:15 Coffee
14:30 Random Graphs and Percolation Theory
15:45 Break
16:00 Connectivity and Coverage

Friday
8:50 Coffee
9:00 Multihop Analysis of Poisson Networks
10:30 Break
10:45 Analysis of General Networks
12:15 Lunch break
13:15 Emerging Architectures
14:45 Coffee
15:00 Modeling and Managing Uncertainty
16:30 Closing

COURSE CONTENT
This short course consists of eight lectures. It assumes a basic
knowledge of probability and wireless communications and is otherwise self-contained. The titles and brief content description of the eight lectures are:

1. Introduction and a Key Result
We motivate the stochastic geometry approach to wireless network
analysis and design and prove a key result on the interference and
outage in networks where nodes are distributed as a Poisson point
process.

2. Throughput Analysis and Design Aspects
Starting from the key result derived in Lecture 1, we derive throughput
results for Poisson networks and discuss the effects of power control,
MAC schemes, spread-spectrum schemes, and interference cancellation. We also give a geometric interpretation of fading.

3. Random Graphs and Percolation Theory
Here we define the disk graph and other random geometric graphs to model connectivity properties of wireless networks, and we give an
introduction to percolation theory on lattices and Poisson networks. We discuss several applications, including a geometric approach to
incorporate secrecy constraints.

4. Connectivity and Coverage
We discuss necessary and sufficient conditions for network connectivity and coverage in sensor networks.

5. Multihop Analysis of Poisson Networks
We extend the analysis from Lectures 1 and 2 to the end-to-end case.
This includes optimal routing, the notion of local delay, and
interference correlation. We also present fundamental bounds on how fast information can spread in Poisson networks.

6. Analysis of General Networks
We introduce more general tools from stochastic geometry that permit the analysis of networks whose nodes are not Poisson distributed. We first focus on clustered networks and then present results for the general case.

7. Emerging Architectures
We apply the analysis techniques introduced in the previous lectures to cognitive networks, cellular systems with femtocells, and cellular
systems with relaying.

8. Modeling and Managing Uncertainty
The final lecture puts the course material in a broader context and
discusses general issues including the question whether wireless systems are or should be interference-limited and node mobility. We also give an outlook of current and future directions and open problems in this area of wireless networks analysis and design using stochastic geometry.