Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.594103
Title: Digital baseband modelling and predistortion of radio over fibre links
Author: Vieira, Luis Carlos
Awarding Body: University of Kent
Current Institution: University of Kent
Date of Award: 2012
Availability of Full Text:
Access from EThOS:
Abstract:
Radio-over-Fibre (RoF) is an alternative solution for the distribution of RF communication signals between a central base station and remote antenna units (RAUs) and can be seen as an enabling · technique to increase the coverage area of broadband wireless access networks. However, the nonlinear distortion introduced by the ReF link may impose serious limitations on the system performance, especial ly considering the high peak-Ie-average power ratio PAPR of complex signal formats, such as orthogonal frequency-division multiplexing (OFDM) and Widehand Code Division Multiple Access (WCD:MA). The objectives of this research are to study and model the nonlinearities of ReF links and to investigate a predistortion approach for this type of link, considering an environment of broadband wireless services. A baseband-behavioural-based modelling approach has been adopted for the nonlinear modelling studies of RoF links and the performance of an off- line training digital predistortion approach to RoF links has been experimentally investigated. The application of memory polynomials (MPs) to the baseband behavioural modelling of both nonlinear and memory effects of directly modulated RoF links has been reported. The robustness of this model as a baseband representation of nonlinear RoF links was demonstrated considering different operating conditions and model parameters. with excellent modelling accuracy reported for most of the experimental cases. The performance of a memory-po lynomial-based digital predistortion technique fo r OFDM-RF RoF links has been investigated within the RF frequency range from 1.85 to 3.7 GHz, up to 20 link lengths. For example, the spectral regrowth was reduced by around 17.5 dB and the = error vector magnitude (EVM) from 8.4 % to 1.4 % for a 2A7·GHz link case. This work represents the first experimental demonstration of the memory-polynomial·based digital predistortion approach to OFDM-RF RoF links using an off-line least-squares-based identification method.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.594103  DOI: Not available
Share: