Date of Award

12-2014

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Computer Engineering and Sciences

First Advisor

Ivica Kostanic

Second Advisor

Susan Earles

Third Advisor

Brian Lail

Fourth Advisor

Muzaffar Shaikh

Abstract

Relays play important role in deployment of Long Term Evolution (LTE) and LTE-Advanced systems. They are used to enhance coverage, throughput and system capacity. The impact of the relay deployment on the network planning has to be modeled so that the above mentioned goals can be achieved. For successful modeling one needs to predict the path loss on the link between an eNodeB and a relay station. The review of literature shows that there is a general shortage of measured data to help empirical understanding of the propagation conditions in relay environment. This dissertation aims at developing empirical path loss models for relaying systems operating in 1900 MHz frequency band. The path loss models are derived on a basis of an extensive measurement campaign conducted in a typical suburban environment. The path loss modeling takes into account the impact of the relay antenna height and therefore, an antenna height correction factor is derived and included in the modeling. The parameters of iv those empirical models include the slope (m) and the intercept (PL0) for each level of the examined relay heights. The parameters are determined from empirical studies and through the appropriate linear regression process. In addition to path loss modeling, the validity of some commonly used propagation models is evaluated by comparing their predictions to the measurements. It was found that the path loss may be modeled successfully with a slightly modified log-distance propagation model. Two new different approaches for development of the statistical propagation path loss models for relay systems has been accomplished. The first approach includes fixed slopes and intercepts along with a distance dependent antenna height correction. The second approach does not include the relay antenna height correction factor but the height of the relay antenna becomes part of the slope and intercept calculation. For both approaches, the results show that model predictions are in good agreement with the measurement. As far as the validity of the commonly used models is concerned, the comparison to the measurements reveal that the applicability of those models for relaying environment is still debatable. Some modifications are introduced in order to improve the performance of these models. With the proposed correction factors all the examined models perform adequately and with approximately the same accuracy

Share

COinS