MONAMI 2011, 21-23 September 2011, Aveiro, Portugal
MONAMI 2010

Tutorial

Wednesday 21 September 2011, 9:00 - 13:00

Cooperative Wireless Networks: From Theory to Practice

Stefan Valentin
Bell Labs, Alcatel-Lucent Deutschland AG, Germany
stefan(dot)valentin(at)alcatel-lucent(dot)com

Hermann S. Lichte
net mobile AG, Germany
hermann(dot)lichte(at)net-m(dot)de



Download MONAMI 2011 Tutorial Slides

Reference: S. Valentin and H. S. Lichte, "Cooperative Wireless Networks: From Theory to Practice", Tutorial in the Third International ICST Conference on Mobile Networks And Management (MONAMI'11), Aveiro, Portugal, 21-23 September, 2011.


Abstract

Currently, industry and academics seeks to understand when cooperative wireless networks perform best, how the performance of cooperative techniques degrades under practical assumptions, and which gains remain if cooperation is integrated into a full wireless system. At the moment, these open questions are a major roadblock in standardization and system design. By answering them, this tutorial provides researchers and engineers with the background and tools to bring cooperative networks into practice.

Keywords

cooperative communication, cooperative relaying, opportunistic routing, collaborative multipoint, prototyping and testbeds

Contributions

This tutorial bridges the gap between analyzing and prototyping cooperative communication. Based on recent advances in information-theoretical models and metrics, we will discuss how closely analytically predicted gains are achieved by current prototypes. In particular, we will (i) introduce theoretical tools to assess the performance of ideal cooperative networks even under realistic constraints, (ii) identify critical system functions that largely affect cooperative gains, and (iii) describe how these functions can be modeled and implemented such that the benefit of cooperation remains even in realistic scenarios.

Finally, we provide an extensive survey on current testbeds for cooperative networks. We describe prototypes from industry and academics in detail and study how closely the presented theoretical methods capture experimental results. To our audience, this tutorial not only provides the theoretical background and tools to critically assess the real-life performance of upcoming cooperative networks but also summarizes starting points and open issues to implement these.

Audience

This tutorial is designed for researchers who want to evaluate cooperative systems without having to read the variety of papers in the field. Our tutorial will also be extremely useful for engineers involved in prototyping or testing cooperative networks that want to establish a theoretical background or learn from the experience with earlier testbeds. Also research students who look for open challenges in theoretical and experimental evaluation and managers who want to critically assess the value of upcoming cooperative systems will profit from this tutorial. The level of treatment is in-depth but suitable for an audience with modest background in wireless communications. Specific knowledge in information theory, cooperative relaying, or MIMO systems is not required.

Novelty

Unlike other tutorials and textbooks in the field, this tutorial spends only little time to reveal the fundamentals of cooperative communication. Instead, we focus on recent results in theory and practice and provide tools that enable engineers and researchers to evaluate their own systems. We point out critical system aspects by reviewing state-of-the-art prototypes and recent experimental results. So far, this concise transition from theory to practice is neither provided by other tutorials nor by textbooks in the field.

Outline

1. Introduction

This part introduces the concept of cooperative relaying, the main problems in the field, and the tutorial objectives. Particular emphasis is on the current status of relaying systems in theoretical research, experiments and prototyping, as well as in standardization.

2. Technologies

This part starts with a taxonomy of cooperative technologies. These technical criteria (e.g., relay type, forwarding strategy, employed channel knowledge) are then used to detail the fundamental approaches in cooperative relaying. Theoretical performance bounds are discussed where appropriate. Finally, the taxonomy is used to categorize common relaying approaches such as Selection Decode-and-Forward, Coordinated multipoint (CoMP), and Cooperative MIMO.

3. Application

Beneficial scenarios are discussed for a variety of relaying techniques. In particular, it is discussed which relaying techniques suits best to cope with shadowing and mobility or to improve capacity and coverage. We highlight that there is not a single "best" technique to succeed in each scenario. Hence, for each technique, we explain its specific operation region based on the theoretical origin of the gains (like spatial diversity or multiplexing) and contrast with the strongest competitors for relaying in a given scenario (e.g., Hybrid ARQ can be used instead of relaying to cope with mobility). This systematic overview helps our auditorium to apply the right cooperation technique to a given scenario.

4. Theory

This large part comprehensively describes the theory behind cooperative communication. It starts by reviewing fundamental constraints (like a system-wide energy limit and half-duplex operation) that have to be accounted for when analyzing relaying systems. Then the theory behind the information-theoretical relay channel is described. Besides classic results, very recent theoretical findings on the capacity for arbitrary relaying systems are revealed.

The majority of this part is devoted to theoretical tools to analyze cooperative relaying systems. To this end, we describe cut-set analysis, outage analysis, and several methods to approximate the capacity and error rate of cooperative networks. The conditions to apply these tools are discussed and various examples are given. All in all, this part should not only provide an overview on the theoretical bounds but should enable our auditorium to analyze the performance of their own systems.

5. Practice

The second major part of our tutorial focuses on the challenges when bringing cooperative communication into practice. The first challenge is how to model realistic effects such as limited channel knowledge, realistic combining, or synchronization. Known and new models are discussed and connected to the theoretical part of this tutorial.

The second challenge is to integrate relaying into current systems. To this end, we describe necessary system changes to solve new problems caused by relaying at the Medium Access Control (MAC) and Physical (PHY) layer. For instance, we describe practical combining schemes, mechanisms at the \ac{MAC} layer to initialize cooperation and to solve relay blocking, as well as techniques to synchronize cooperative transmitters.

Moving closer to implementation, we provide an overview of prototyping platforms for relaying. By systematically discussing their pros and cons, we help our auditorium to choose the right platform for their prototyping project.

Finally, we summarize the current results from experiments and field measurements. Besides describing the experimental setups (such as in Figure 1.), we review how measurement results reflect the theoretical findings and technologies presented above. This consistently shows which theoretical statements are supported by experiments, which of them still need to be justified, and which practical problems still need to be solved.


Figure 1. One of the discussed testbeds: A test train carries two software-defined radios which operate as source s and relay r to cooperatively reach a third station d beside the track. The devices can fall back to standard WLAN operation and reach full IEEE 802.11a/g rates.

6. Conclusion and Discussion

We conclude our tutorial by reviewing which techniques for cooperative communication are mature and which techniques still need theoretical and practical justification. Then we summarize application scenarios and open research problems such as pairing and interference control from an academic and industry perspective.

Finally, we discuss the future of cooperative communication. Two basic cases (capacity, coverage) are compared and a new use case (relaying as support technology) is revealed. The discussion is completed by a financial perspective and by a list of recommended literature.

7. Appendix

In addition to the recommended literature, the full list of references as well as proofs and derivations are provided in the appendix.

Biographies

Stefan Valentin studied communication technology and electrical engineering at the TU Berlin. Since 2005 he served as a research and teaching associate at the University of Paderborn (Germany) where he received his PhD in Computer Science with "summa cum laude" in 2010. Stefan was invited lecturer at the International Centre of Theoretical Physics (Trieste, Italy) and invited scientist at the Carleton University (Ottawa, Canada). In 2010, he joined Bell Labs in Stuttgart, Germany where he works on cooperative communication and context-aware resource allocation.

Hermann S. Lichte received his PhD in Computer Science from the University of Paderborn in 2011 with "magna cum laude". In his research he investigated the costs of cooperative relaying in wireless multi-hop networks. Hermann now works as an Innovation Manager at net mobile AG in Germany, where he is developing state-of-the-art e-commerce solutions for mobile networks.

Both authors have published widely in the field of cooperative communication, won the ACM SIMUTools best paper award in 2008, and have a remarkable experience in testbed development. Since 2005, Stefan supervised five academic and industrial prototyping projects as well as extensive field tests for cooperative relaying systems and holds several patents in this field. Hermann, has three years experience in prototyping cooperative relaying on Software-defined Radios (SDRs) and on off-the-shelf WLAN adapters. In 2008, both authors demonstrated the world's first cooperative WLAN prototype that reached full IEEE 802.11a/g rates.