Dissertation on Advanced Radio Resource Management for Relaying in Heterogeneous Mobile Networks
On September 16, 2015, Thiago Martins de Moraes successfully defended his Ph.D. thesis Advanced Radio Resource Management for Relaying in Heterogeneous Mobile Networks. The Ph.D. examination committee was formed by Prof. Marko Lindner (Institute of Mathematics) and the two reviewers Prof. Gerhard Bauch and Prof. Wolfgang Kellerer (TUM Technische Universität München).
Abstract
The past decade has been characterized by a drastic change in the mobile user’s behavior. They have migrated from voice and low data rate services to services with huge data rate requirements.
While analyzing those changes, 3rd Generation Partnership Program (3GPP) has predicted that its Universal Mobile Telecommunications System (UMTS) would not accommodate the current data growth for a long time. Hence, they have decided to evolve it and to cope with much higher data rates. The new system was baptized Long Term Evolution (LTE). LTE is a broadband radio access technology, which was first standardized in the last quarter of 2008. A highly adaptive physical layer was designed, which allows a great number of users and an efficient usage of the radio link. Initially, LTE could achieve a data rate of 300 Mbps in the downlink and 75 Mbps in the uplink.
Although LTE has opened several new service possibilities, it was still under the goals established for the next mobile service generation. Hence, new features such as relays, carrier aggregation, and new MIMO options were introduced to LTE. These improvements were included in the so-called LTE-Advanced.
Relays are a very attractive feature. The main task of these nodes is to boost the signal quality of the users located in bad signal spots. They are cheap and low power nodes, which decode the base station messages and send to the user a re-encoded version of them with better signal quality. They are an interesting manner of solving the coverage-hole problem created in the early deployment phase. Coverage-holes are created because of the low number of base stations (eNBs) in this early stage: the entire service area might not be covered and the relays would act as replacements for the base stations.
Relays can only serve users if the eNB sends data to them in the radio interface. Resource partition is a process in which the eNB decides how many resources are reserved in the air interface to serve the relays. Using a proposed algorithm, the Fair Throughput, we were able to demonstrate that the relays can improve the signal of the users in the cell-edge without degrading the services of the users in the cell center. This algorithm consists in averaging the cell throughput and allocating resources to the relay such that the relayed users can achieve the cell average throughput.
Using these results, we have analyzed scheduling of QoS-aware services in the presence of relays. We have proposed a frequency-time scheduler, which manages to improve the service quality of the cell-edge users. Figure-1 shows a comparison between the deployment without (right) and with (left) relays. Without the relays, eNB provides a low service quality to the cell-edge users while using the relays the users achieve a much better service quality.
