Ultra Wideband (UWB) communication systems occupy large bandwidths with very low power spectral densities. This feature makes UWB channels highly rich in multipaths. To exploit the temporal diversity, a UWB receiver usually incorporates Rake reception. Each multipath in the channel carries just a fraction of the signal energy. This phenomenon dictates a Rake receiver with a large number of fingers to achieve good energy capture and output signal to noise ratio (SNR). Eventually, the Rake structure becomes very complex from analysis and design perspectives and incurs higher manufacturing cost. The first contribution of this thesis is to propose channel shortening or time domain equalization as a technique to reduce the complexity of the UWB Rake receiver. It is analyzed that most of the existing channel shortening equalizer (CSE) designs are either system specific or optimize a parameter not critical or even available in UWB systems. The CSE designs which are more generic and use commonly critical cost functions may perform poorly due to particular UWB channel profiles and related statistical properties. Consequently, the main contribution of the thesis is to propose several CSE designs to address the specific needs of UWB systems. These CSE designs not only exploit some general but also some UWB specific features to perform the task more efficiently. The comparative analysis of the proposed CSEs, some existing designs and the conventional Rake structures leads towards the conclusion. It is finally shown that the use of CSE at the receiver front end greatly simplifies the Rake structure and the associated signal processing.

Ultra Wideband (UWB) communication systems occupy large bandwidths with very low power spectral densities. This feature makes UWB channels highly rich in multipaths. To exploit the temporal diversity, a UWB receiver usually incorporates Rake reception. Each multipath in the channel carries just a fraction of the signal energy. This phenomenon dictates a Rake receiver with a large number of fingers to achieve good energy capture and output signal to noise ratio (SNR). Eventually, the Rake structure becomes very complex from analysis and design perspectives and incurs higher manufacturing cost. The first contribution of this thesis is to propose channel shortening or time domain equalization as a technique to reduce the complexity of the UWB Rake receiver. It is analyzed that most of the existing channel shortening equalizer (CSE) designs are either system specific or optimize a parameter not critical or even available in UWB systems. The CSE designs which are more generic and use commonly critical cost functions may perform poorly due to particular UWB channel profiles and related statistical properties. Consequently, the main contribution of the thesis is to propose several CSE designs to address the specific needs of UWB systems. These CSE designs not only exploit some general but also some UWB specific features to perform the task more efficiently. The comparative analysis of the proposed CSEs, some existing designs and the conventional Rake structures leads towards the conclusion. It is finally shown that the use of CSE at the receiver front end greatly simplifies the Rake structure and the associated signal processing.

Le contexte de ces travaux de thèse est la transmission dite faster-than-Nyquist (FTN). Cette technique propose d’augmenter l’efficacité spectrale en augmentant lerythme de transmission au-delà de la bande occupée par le signal émis, indépendamment de laconstellation choisie. Il a été montré que le FTN offre des taux d’information supérieurs à ceuxdes systèmes de Nyquist. Toutefois, le non respect du critère de Nyquist entraîne l’apparitiond’interférence entre symboles et des techniques de réception appropriées doivent être utilisées.La technique de réception dite channel shortening consiste à filtrer la séquence reçue puis àcalculer des probabilités symbole a posteriori approximatives à l’aide de l’algorithme BCJRen considérant une réponse de canal modifiée, de longueur réduite. Dans la littérature, enprésence d’information a priori, les filtres du récepteur channel shortening sont optimiséssous critère de maximisation de l’information mutuelle généralisée (IMG) en utilisant desméthodes numériques. Nous proposons dans ces travaux de thèse une solution analytiquepour l’ensemble des filtres channel shortening sous critère de maximisation de l’IMG lorsquele récepteur dispose d’information a priori. Nous démontrons ensuite que l’égaliseur au sens dela minimisation de l’erreur quadratique moyenne (MMSE) est un cas particulier de l’égaliseurchannel shortening. Dans le cadre de la turbo égalisation, nous étudions ensuite un estimateurpermettant d’obtenir l’information a priori à partir de l’information en sortie du décodeurcorrecteur d’erreurs. Finalement, nous évaluons les performances du système complet aveccodage correcteur d’erreurs sur canal à bruit additif blanc Gaussien. / In order to increase the spectral efficiency of digital communications systems,the faster-than-Nyquist (FTN) approach increases the symbol rate beyond the occupied bandwidthof the transmitted signal independently of the constellation type and size. It has beenshown that information rates of FTN systems are greater than those of Nyquist systems.However, the non-compliance of the Nyquist criterion causes inter-symbol interference to appearand therefore appropriate reception techniques must be used. At reception, the channelshortening approach consists on a receiving filter followed by a BCJR algorithm computingapproximate a posteriori symbol probabilities by considering a modified channel response ofreduced length. In the literature, the channel shortening receiving filters are chosen to maximizethe generalized mutual information (GMI). Such optimization is performed by usingnumerical optimization methods. In this PhD thesis, we propose a closed-form solution forall channel shortening filters considering the GMI maximization criterion. We show that theminimum mean square error (MMSE) equalizer is a particular case of the channel shorteningapproach. Within the frame of turbo equalization, we then study a suitable estimator allowingto obtain symbols a priori information from the information provided by the a decoder. Finally,we study the performance of the complete system with channel coding over an additivewhite Gaussian noise channel.

Considering the communication systems with cyclic prefix (CP), such as orthogonal frequency-division multiplexing (OFDM) modulation and single-carrier cyclic prefixed (SCCP) modulation, when the length of CP is longer than the channel length, the use of cyclic prefix (CP) does not only eliminate the inter-block interference, but also convert linear convolution of the transmitted signal with the channel into circular convolution. Unfortunately, the use of CP significantly decreases the bandwidth utilization. Therefore, to reduce the length of CP is a critical issue. The thesis investigates that how to design a channel-shortening equalizer (CSE) at receiver which forces the length of the effective channel response as short as the CP length. The thesis describes the signal model as a matrix form. The effect channel response after CSE is investigated and then the coefficient of channel shortening filter is obtained using singular value decomposition method under various criterions. We further propose a novel CSE maximizing the shortening signal-to-interference ratio. In addition, it is demonstrated that the proposed CSE has the same performance as the conventional scheme but a lower computation complexity.