Channel estimation and signal-space diversity for vector-valued transmissions [Elektronische Ressource] / von Shawki Abdelfattah Ahmed Saad Abdelkader

Informations

• Publié par: universitat_ulm
• Publié le: 01 janvier 2007
• Nombre de lectures: 26
• Langue: English
• Poids de l'ouvrage: 2 Mo

Extrait

Channel Estimation and Signal-Space
Diversity for Vector-Valued Transmissions
DISSERTATION
zur Erlangung des akademischen Grades eines
Doktor{Ingenieurs
(Dr.{Ing.)
der Fakult at fur Ingenieurwissenschaften und Informatik
der Universit at Ulm
von
SHAWKI ABDELFATTAH AHMED SAAD ABDELKADER
AUS KAIRO, AGYPTEN
1. Gutachter: Prof. Dr.-Ing. Jurgen Lindner
2.hter: Prof. Hans-J org P eiderer
Amtierender Dekan: Prof. Dr. rer. nat. Helmuth Partsch
Ulm, 31. Januar 2007Acknowledgements
Praise to God, the most gracious and the most merciful. Without His blessing and
guidance, my accomplishments would have never been possible.
I would like to thank all who helped me during the course of this work. My words will
fail to express my deepest heartfelt thanks to my supervisor, Prof. Dr.-Ing. Jurgen Lindner,
for giving me the opportunity to carry out this thesis and for his guidance, support, motivation
1)and encouragement . Without his continuous support and patience this work would not have
been possible. I thank him also for his concern and assistance even with other things in my
life.
I also would like to take this opportunity to extend my thanks to all my colleagues working
in the Institute for their truthful and sincere concerns about my study and myself and for the
favorable working environment. I would like to particularly acknowledge the help of Dr. rer.
nat. Werner Teich, and I would like to express my special thank to Doris Y. Yacoub, Christian
Pietsch, Ivan Perisa and Markus Dangl for their friendly support. A special thank is due to my
parents who gave me always support during this work. Finally, and most important, I should
mention that I probably would never have nished this thesis at all without the persuasion of
my wife Heba.
1)The disseration report was concluded within my research period in the Institute of Information Technology
University of Ulm
VContents
1 Introduction and Motivation 1
2 Fundamentals 5
2.1 Conventions of notations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 MIMO channel identi cation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2.1 Rayleigh fading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2.2 MIMO channel model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.2.3 Linear algorithms for MIMO channel identi cation . . . . . . . . . . . . 10
2.2.4 General space model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3 MMSE for MIMO channel estimation . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3.1 Approximate MMSE estimator . . . . . . . . . . . . . . . . . . . . . . . 13
2.4 Detection based on the estimated channel . . . . . . . . . . . . . . . . . . . . . 13
2.5 Signal-space diversity and fading e ects . . . . . . . . . . . . . . . . . . . . . . . 15
2.5.1 Signal-space diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.5.2 Lattice constellations and optimum design . . . . . . . . . . . . . . . . . 17
2.5.3 Multidimensional constellations design . . . . . . . . . . . . . . . . . . . 18
2.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3 MIMO Channel Estimation in the Presence of Errors 21
3.1 Channel capacity with imperfect channel knowledge . . . . . . . . . . . . . . . . 22
3.1.1 Lower bound of the capacity with erroneous estimated channel . . . . . . 22
3.2 The in uence of channel estimation on optimum and suboptimum receiver . . . 24
VIIContents
3.2.1 Modelling channel estimation errors . . . . . . . . . . . . . . . . . . . . . 26
3.2.2 Cramer-Rao bound as a quality measure for channel estimation . . . . . 27
3.3 Diversity combining with channel estimation errors . . . . . . . . . . . . . . . . 28
3.3.1 Optimal combining and its error performance . . . . . . . . . . . . . . . 28
3.3.2 Suboptimal combining and its error performance . . . . . . . . . . . . . . 31
3.4 MIMO with signal-space diversity in the presence of channel errors . . . . . . . 32
3.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4 Adopting Signal-Space Diversity to Combat Channel E ects 39
4.1 From Rayleigh fading to Gaussian channel . . . . . . . . . . . . . . . . . . . . . 40
4.1.1 High diversity order rotated constellation techniques . . . . . . . . . . . 41
4.1.2 Unequal values of the rotation matrix element weights . . . . . . . . . . 44
4.1.3 Upper bounds of signal-space diversity . . . . . . . . . . . . . . . . . . . 44
4.2 The construction of the rotation matrix . . . . . . . . . . . . . . . . . . . . . . . 45
4.2.1 Hadamard and Fourier spreading transforms . . . . . . . . . . . . . . . . 45
4.2.2 Rotation using rotated spreading . . . . . . . . . . . . . . . . 46
4.3 Signal-space diversity - DAST . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.3.1 DAST codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.3.2 Maximum achievable rate of the rotated constellation . . . . . . . . . . . 48
4.3.3 Constructing DAST codes . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.3.4 DAST codes and combining technique e ect . . . . . . . . . . . . . . . . 51
4.4 The similarity between DAST and MC-CDM . . . . . . . . . . . . . . . . . . . . 53
4.5 Simulation results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.5.1 Half diversity-order of signal space . . . . . . . . . . . . . . . . . . . . . 60
4.5.2 Full diversity-order of signal space . . . . . . . . . . . . . . . . . . . . . . 62
4.5.3 Signal-space with space-time diversity techniques . . . . . . . . . . . . . 63
4.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
VIIIContents
5 Recursive Algorithms for MIMO Channel Estimation 75
5.1 Adjustable time-varying channel model . . . . . . . . . . . . . . . . . . . . . . . 76
5.1.1 Clarke’s channel model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.1.2 Pop&Beaulieu’s channel model . . . . . . . . . . . . . . . . . . . . . . . 78
5.1.3 Modi ed-P op&Beaulieu’s channel model . . . . . . . . . . . . . . . . . . 79
5.1.4 Establishing Rayleigh time-varying channel model . . . . . . . . . . . . . 80
5.2 Channel estimation and Alamouti Scheme . . . . . . . . . . . . . . . . . . . . . 80
5.2.1 Alamouti’s space-time block coding scheme . . . . . . . . . . . . . . . . . 83
5.2.2 STBCs from complex notations to a linear (real) transformation . . . . . 85
5.3 LS and linear-MMSE channel estimation . . . . . . . . . . . . . . . . . . . . . . 88
5.3.1 Constructing the correlation matrices . . . . . . . . . . . . . . . . . . . . 89
5.4 LMS and normalized-LMS algorithms for MIMO-channel estimation . . . . . . . 90
5.4.1 LMS channel estimator . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
5.4.2 Normalized-LMS for channel estimation . . . . . . . . . . . . . . . . . . 93
5.5 Channel estimation using RLS and adaptive--RLS algorithms . . . . . . . . . . 94
5.5.1 RLS for channel estimation . . . . . . . . . . . . . . . . . . . . . . . . . 94
5.5.2 Adaptive forgetting factor for RLS algorithm . . . . . . . . . . . . . . . . 95
5.6 Simulation results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
5.6.1 Channel estimation and Alamouti scheme . . . . . . . . . . . . . . . . . 96
5.6.2 Channel for DAST codes . . . . . . . . . . . . . . . . . . . . . 102
5.6.3 DAST in the presence of channel estimation errors . . . . . . . . . . . . . 105
5.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
6 Summary, Conclusion, and Suggestions for Future Work 113
6.1 Summary and conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
6.2 Suggestions for future work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
A List of Abbreviations 117
Bibliography 119
IXList of Figures
2.1 Rotated constellation in two dimensional with BPSK modulation is an example
of the signal-space diversity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2 General Block diagram of vector-valued transmission system using signal-space
diversity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.1 Transmission vector blocks generally contain pilot and data symbols slots . . . . 26
3.2 Theoretical capacities for MIMO (82) system with di eren t average percentages
of channel estimation errors (Equation (3.16)). . . . . . . . . . . . . . . . . . . . 36
4.1 Transmission model on symbol basis for MIMO system using general coding ma-
trix containing signal-space and space-time matrices. The matched lter matrix
in general maximizes form the SNR at the receiving end. . . . . . . . . . . . . . 54
4.2 Rotated constellations of BPSK of di eren t diversity orders. Half diversity order
is represented by Boutros 1,2, and 3 in addition to full diversity order whichted by Bury, Damen and UEW. . . . . . . . . . . . . . . . . . . . . . . 59
4.3 BPSK and QPSK are rotated (spread) by using Hadamard matrix. The diversity
order is 8 and "Diversity8" represents the diversity upper bound. . . . . . . . . 60
4.4 Di eren t half diversity order (Boutros1, 2, and 3) rotation angles. The rotation
matrix is the the rotated Hadamard matrix with diversity 8 and data constella-
tions are BPSK. . . . . . . .