ETD

Archivio digitale delle tesi discusse presso l'Università di Pisa

Tesi etd-03192008-222619


Tipo di tesi
Tesi di dottorato di ricerca
Autore
CARTA, MICHELE
URN
etd-03192008-222619
Titolo
FREQUENCY RECOVERY AND RECEIVER DESIGN ISSUES IN FBMC WIRELESS SYSTEMS
Settore scientifico disciplinare
ING-INF/03
Corso di studi
INGEGNERIA DELL'INFORMAZIONE
Relatori
Relatore Prof. Mengali, Umberto
Relatore Prof. Reggiannini, Ruggero
Relatore Ing. Lottici, Vincenzo
Parole chiave
  • receiver design
  • FilterBank MultiCarrier (FBMC) modulation
  • doubly selective channels
  • Carrier Frequency Offset (CFO) recovery
  • TErrestrial Trunked RAdio (TETRA)
Data inizio appello
10/06/2008
Consultabilità
Completa
Riassunto
This Ph.D. dissertation deals with one of the most prominent example of MC
modulation, Filter Bank MultiCarrier (FBMC) modulation, and aims at reporting
the main results of our research activity carried out over the last three years in
the field of FBMC wireless systems.

The first chapter introduces the basic concepts of wireless communication
channels and multicarrier modulation. In the first part, FBMC modulation is presented.
We provide a detailed description of the modulator and demodulator architecture which
are based on digital signal processing techniques. The last section is devoted to
an overview of the main physical aspects of the recent TEDS ETSI standard for
PMR/PAMR communications, which relies on the FBMC technique.

The core topic of the second chapter is carrier frequency offset recovery (CFO) for FBMC
transmission over time-frequency selective fading channels. After a brief statement
of the problem, we analyse three different frequency estimation algorithms
that arise from the maximum likelihood principle. Specifically the schemes
exhibit an open-loop structure, intended for burst transmissions, and are all based
on the use of known pilot symbols scattered across the transmitted burst.
Performance analysis quantifies the accuracy of the proposed algorithms in typical
mobile wireless scenarios, showing they outperform conventional NDA frequency
recovery even combined with a lower computational complexity.

The third chapter deals with receiver design issues for FBMC burst transmission over
doubly selective fading channels with emphasis on synchronization and channel
estimation aspects. Specifically, a ML-derived symbol timing recovery
algorithm is proposed that relies on the transmission of a short known preamble.
The carrier synchronization task is implicitly carried out by a properly
modified Bayesian channel estimation algorithm that exploits pilot symbols uniformly
spaced throughout the subcarriers. The proposed receiver architecture is thoroughly
discussed and the schemes are assessed, evaluating their impact on the
receiver performance in terms of FER in typical mobile wireless channel conditions.
File