Constellation Optimization in the Presence of Strong Phase Noise
Artikel i vetenskaplig tidskrift, 2013

In this paper, we address the problem of optimizing signal constellations for strong phase noise. The problem is investigated by considering three optimization formulations, which provide an analytical framework for constellation design. In the first formulation, we seek to design constellations that minimize the symbol error probability (SEP) for an approximate ML detector in the presence of phase noise. In the second formulation, we optimize constellations in terms of mutual information (MI) for the effective discrete channel consisting of phase noise, additive white Gaussian noise, and the approximate ML detector. To this end, we derive the MI of this discrete channel. Finally, we optimize constellations in terms of the MI for the phase noise channel. We give two analytical characterizations of the MI of this channel, which are shown to be accurate for a wide range of signal-to-noise ratios and phase noise variances. For each formulation, we present a detailed analysis of the optimal constellations and their performance in the presence of strong phase noise. We show that the optimal constellations significantly outperform conventional constellations and those proposed in the literature in terms of SEP, error floors, and MI.

maximum likelihood (ML) detection

Constellations

phase noise

mutual information

symbol error probability

Författare

Rajet Krishnan

Chalmers, Signaler och system, Kommunikation, Antenner och Optiska Nätverk

Alexandre Graell i Amat

Chalmers, Signaler och system, Kommunikation, Antenner och Optiska Nätverk

Thomas Eriksson

Chalmers, Signaler och system, Kommunikation, Antenner och Optiska Nätverk

Giulio Colavolpe

Universita degli Studi di Parma

IEEE Transactions on Communications

0090-6778 (ISSN) 15580857 (eISSN)

Vol. 61 12 5056-5066 6648358

Styrkeområden

Informations- och kommunikationsteknik

Ämneskategorier

Elektroteknik och elektronik

DOI

10.1109/TCOMM.2013.102313.130131

Mer information

Senast uppdaterat

2018-03-19