Skip to Main content Skip to Navigation
Theses

Techniques de démodulation aveugle en interception de signaux MIMO

Abstract : The MIMO (Multiple Input Multiple Output) systems were proposed by the Bell labs in order to increase the capacity of the wireless communications. Then, these systems have been used to improve the reliability of data-transfer by using codes which introduce redundancy. This thesis is achieved in collaboration between the IPSIS-IT Link Company and the SCEE team of Supélec member of the IETR laboratory. The objective of this thesis is to propose blind estimation techniques of the symbols transmitted on MIMO channels. The blind source separation (BSS), allows to estimate in blind way, i.e. without knowledge of the channel and training sequences, the transmitted symbols. This type of method will be used in this thesis mainly. However, the BSS estimates the sources with a certain permutation and rotation. Moreover, the BSS methods using a stochastic gradient descent are slow to converge and with time varying environments they are less accurate. Thus, the subjects tackle in this thesis are: -The exploitation of the redundancy introduced by the STBC coding to raise the ambiguity introduced by the BSS. - The implementation of criteria in analytical way in order to obtain a fast convergence. - The initialization of the Kalman filter by the blind sources separation. This type of association allows to track convolutive or instantaneous channels which vary in time.
Complete list of metadatas

Cited literature [56 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00441830
Contributor : Myriam Andrieux <>
Submitted on : Thursday, December 17, 2009 - 2:12:05 PM
Last modification on : Thursday, January 7, 2021 - 4:13:30 PM
Long-term archiving on: : Saturday, November 26, 2016 - 4:44:41 PM

Identifiers

  • HAL Id : tel-00441830, version 1

Citation

Steredenn Daumont. Techniques de démodulation aveugle en interception de signaux MIMO. Traitement du signal et de l'image [eess.SP]. Université Rennes 1, 2009. Français. ⟨tel-00441830⟩

Share

Metrics

Record views

669

Files downloads

2419