Cooperative Relaying and Adaptive Scheduling for Low Mobility Wireless Access Networks


Ingmar Hammerström


PhD thesis, ETH Zürich, No. 16775, Logos Verlag Berlin, ISBN 978-3-8325-1466-2, pp. 243, 2006.

[BibTeX, LaTeX, and HTML Reference]


In this thesis signal transmission schemes, in particular cooperative relaying schemes, for low mobility wireless networks are developed and analyzed. Low mobility limits the amount of the variation of the physical radio channel over the time. Thus, receiving nodes move only slowly out of an area associated to bad channel conditions, where signal fading is very strong. The use of diversity in the spatial and temporal dimensions mitigates the effects of fading and therefore increases the reliability of radio links in wireless networks. Diversity gains can also be realized in a distributed fashion (cooperative diversity). The main idea of these cooperative diversity schemes is to use relay nodes as virtual antennas to facilitate the communication of one source-destination pair. Relays can be classified as either decode-and-forward (DF) or amplify-and-forward (AF) relays. AF relays, which are considered mainly in this thesis, only retransmit an amplified version of their received signals. This leads to low-complexity relay transceivers and lower power consumption since there is no need of signal processing for decoding. One further advantage of AF relays is that they are transparent to adaptive modulation techniques which may be employed by the source.

First we analyze signaling schemes for one source-destination pair. Several AF relays, which operate within the same frequency band, facilitate the communication of this pair. The signaling schemes depend on the amount of channel state information (CSI) which is available at the relays. If, e.g., the relays know their corresponding first and second hop channel coefficient perfectly, the optimal signaling scheme results in coherent combining of all signal contributions at the destination. An additional requirement for coherent combining of distributed transmitters is a global phase reference, i.e., all local oscillators have to be phase synchronous. If there is no second hop CSI available at the relays one way to increase the reliability of the link is to use transmit diversity techniques. We present a new scheme which is able to achieve full diversity gain at the destination. This scheme essentially translates the block fading time-invariant channel into a time-variant channel by introducing, e.g., time-variant and relay-dependent phase offsets at the relays. This artificial time-variant channel can be exploited by an outer code introduced at the source.

Afterwards, we extend our results to the network-level perspective where multiple source-destination pairs share the same physical resources. A new concept to achieve multiuser diversity gains by opportunistic scheduling in a low mobility cooperative wireless network is presented and analyzed. We refer to this scheme as joint cooperative diversity and scheduling (JCDS). The key idea is to further exploit the artificially introduced time-variance by joint consideration of several source-destination pairs and to schedule, e.g., the link which supports the highest data rate among all links compared to its average supported rate. Note that although fairness of JCDS explicitly benefits from the artificial introduced time-variance it does not assure that each source-destination link is scheduled within specified time frame (short-term temporal fairness). Therefore two further scheduling schemes which are able to guarantee short-temporal fairness while achieving considerable multiuser diversity gains are presented.

Frequency selective broadband channels are addressed, too. We consider a two-hop MIMO-OFDM communication scheme with one source-destination pair and one relay. AF and DF relaying is considered. We examine the possibilities of power allocation (PA) over the subchannels in frequency and space domain to maximize the instantaneous rate of this link if channel state information at the transmitter (CSIT) is available. We consider two approaches: (i) separate optimization of source or relay PA with individual per node transmit power constraints and (ii) joint optimization of source and relay PA with joint transmit power constraint. We show that DF relaying with and without CSIT achieves much higher average rates than AF relaying with CSIT in most of the cases. Moreover, we show that the performance gain due to CSIT compared to no CSIT is much higher in the case of AF relaying than for DF relaying.

In the last part of the thesis we verify the obtained theoretical performance results by means of distributed measurements with the RACooN Laboratory which is available at the Communication Technology Laboratory. We demonstrate in a real-time experiment the proposed cooperative relaying scheme which translates the block-fading environment into a time-variant channel by introducing phase-offsets at the relays. Further, we analyze the performance of the JCDS scheme based on channel measurements which are conducted in two different indoor office environments.

Download this document:


Copyright Notice: © 2006 Logos Verlag Berlin.

This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.