Sie sind hier:
Schriftgröße ändern
Seiteninhalt
Mobile Systems
| Cooperative Diversity in Wireless Networks: Relay Selection and Medium Access - H. Adam |
| Advisor | Univ.-Prof. Dr. Ing. Christian Bettstetter |
| E-Mail | |
| Description | Mobile communication has undergone a tremendous revolution in recent years. The global number of mobile phone subscribers has increased from 750 million to 5.1 billion in the last decade. In developed countries, people have on average more than a single subscription [ITU11]. Beside the subscriber growth, also the achievable data rate of the devices has experienced enormous increases. For instance, Global System for Mobile Communications (GSM) offers a maximum data rate of 9.6 kbit=s, its successor Universal Mobile Telecommunication System (UMTS) reaches data rates up to 384 kbit=s, and Long Term Evolution (LTE) supports 326.4 Mbit=s. We also see an emergence of a network of things, since more and more devices are capable of communicating with each other using the wireless medium. Having more devices which communicate with each other at higher data rates, put high demands on the used technology, especially since the wireless medium is a harsh environment. Mobile communications are affected mostly by a phenomenon called Small Scale Fading, which results in unpredictable and rapid uctuations of the received signal levels. Since the achievable data rate mainly depends on the received signal strength, it is of utmost interest to communication engineers to effectively combat the effects caused by small scale fading. Cooperative diversity is a promising means where wireless communication devices cooperate to mitigate those effects. While the information theory and physical layer aspects of cooperative diversity have been heavily investigated, networking aspects have been mostly unattended. This thesis focuses on the networking aspects of cooperative diversity.
|
|
| Inhomogeneous Node Distribution and Interference in Wireless Networks - U. Schilcher |
| Advisor | Univ.-Prof. Dr.-Ing. Christian Bettstetter |
| E-Mail | |
| Description |
Many research and engineering tasks in the area of wireless communications necessitate simulation based studies of wireless networks. If such simulations should be beneficial, many different aspects of the network and its environment have to be modeled as realistically as possible. The thesis contributes to two groups of such models: the spatial distribution of nodes and the interference model.
When regarding spatial node distributions, much of the research work is based on a uniform distribution. One of the reasons for that is the lack of an easy-to-use model that generates inhomogeneous node distributions. As shown by several researchers, the node distribution has great influence on the performance of different protocols. Hence, the node distribution applied in a simulation study should closely reflect the node distribution of the real scenario, which is rarely a uniform distribution. We propose a model for synthesizing inhomogeneous node distributions, a metric for measuring inhomogeneity, and a mobility model for inhomogeneously distributed mobile nodes to counteract this lack within the research community.
With regard to interference it has to be mentioned that not only the strength of interference, but also its temporal and spatial behavior has strong influence on the performance of many communication protocols. Hence, to analyze these protocols it is necessary to calculate the temporal and spatial correlation of interference. We contribute to this issue by deriving closed-form expressions for the temporal correlation of interference in several scenarios.
Finally, we conduct a simulation based study to compare time and space-time diversity methods with conventional communication protocols. The results of this study show that the performance of a wireless network is to a great extent determined by the interference present in the network. It is therefore advantageous to reduce the interference as much as possible. |
|
| Firefly Synchronization in Wireless Networks - A. Tyrrell |
| Advisor | Univ.-Prof. Dr.-Ing. Christian Bettstetter |
| Description | A striking example of self-organization in nature occurs every evening in some parts of South-East Asia: thousands of fireflies gather on trees at dawn and start emitting flashes regularly; over time, synchronization emerges from a seemingly chaotic situation, which makes it seem as though the whole tree is flashing in perfect synchrony. This fascinating phenomenon is the inspiration for the topic treated in this thesis, which is concerned with slot synchronization in wireless networks.
Synchronization phenomena in nature are mathematically described by the theory of pulsecoupled oscillators (PCOs); each entity naturally oscillates and blinks periodically, and coupling is performed through the discrete emissions of light. Each node adjusts its internal ference when perceiving blinks from its neighbors, and following simple rules, synchronization always emerges after some time. Conditions for convergence under ideal assumptions were derived by Mirollo and Strogatz in their seminal work published in 1990, and provide a framework for the following slot synchronization studies.
The PCO synchronization rules are remarkably simple and robust, which makes their application to wireless networks very appealing. In particular, slot synchronization requires nodes in the network to agree on a common time reference for the start of a slot, in a similar way to fireflies that agree on a common blinking instant. Direct application of the PCO rules is not feasible, and an adaptation, termed Mobile Emergent Firefly Synchronization (MEMFIS), is proposed so that constraints of wireless networks are integrated with the PCO rules. With this modification, the simplicity and robustness of the PCO scheme is retained; nodes are able to synchronize starting from any random misalignment, and achieve an accuracy equal or lower to the direct propagation delay.
Application of the PCO model to cellular systems is investigated. The goal is to maintain base stations synchronized, even when there is no direct communication between them. Synchronization in Cellular Firefly Synchronization (CelFSync) is performed by letting some
selected user terminals participate in the network synchronization process, and achieving an out-of-phase synchronization regime. Furthermore propagation delays, which are problematic in large-scale networks, are mitigated by combining the proposed adaptation with the timing advance procedure, so that an acceptable inter-base station accuracy is achieved.
Alexander Tyrrell's thesis Firefly Synchronization in Wireless Networks is now available. |
|
