Processeur de signal numérique

Un DSP (de l'anglais « Digital Signal Processor », qu'on pourrait traduire par « processeur de signal numérique » ou « traitement numérique de signal ») est un microprocesseur optimisé pour exécuter des applications de traitement numérique du signal (filtrage, extraction de signaux) le plus rapidement possible. Les DSP sont utilisés dans la plupart des applications du traitement numérique du signal en temps réel. On les trouve dans les modems (modem RTC, modem ADSL), les téléphones mobiles, les appareils multimédia (lecteur MP3), les récepteurs GPS… Ils sont également utilisés dans des systèmes vidéo, les chaînes de traitement de son, partout où l'on reçoit un signal complexe que l'on doit modifier à l'aide du filtrage. Au début des années 1990, ils avaient été utilisés dans certains ordinateurs comme le NeXT Cube et l'Atari Falcon. Traitement numérique du signal Le traitement numérique du signal implique la réalisation de nombreuses opérations mathématiques. En compara

Parallel architecture for high-speed analog-to-digital conversion

Guillaume Ding

Nowadays digital signal processing systems used for radar applications, communication systems or RF measurement equipments, require very high sample-rates. Sometimes these sample-rates are beyond the possibilities offered by conventional ADCs. To overcome these limits parallel architectures have been developed. The most commonly used it the one called "time-interleaved" conversion. This technique allows to achieve very-high sample-rates with circuits working at a lower frequency. The accuracy of "time-interleaved" systems is sensitive to sample-time errors. Some calibration techniques have been developed to reduce this sensitivity. They involve very sophisticated digital signal processing and, in most of the cases, they are not directly implemented on silicon but applied on measurement results in software. The goal of this thesis is to study the feasibility of a new parallel architecture for analog-to-digital conversion. This architecture must present a higher robustness to sample-time errors. The first part of this work is dedicated to time-interleaved converter. An analysis of their sensitivity to several imperfections, such as mismatches and systematic and random sample-time error is presented. This analysis is followed by a description of time-interleaved converter evolution, since the first implemented prototype to the current state of the art of the domain. The second part of this thesis focuses on the development of the new conversion technique called "frequency-interleaved". Two different approaches are studied: the first one is based on a Fourier series decomposition of the signal to convert and the second one is based on a Walsh series decomposition. During this study, theoretical and practical aspects are faced the one with the other, to combine signal processing and microelectronics together. It appears that the Fourier series approach offers modest performances and presents serious problems of implementation. Based on this study, the design of functional blocks of a Walsh series based system is proposed.

EPFL2009

Exploring information retrieval using image sparse representations

William Guicquéro

New advances in the field of image sensors (especially in CMOS technology) tend to question the conventional methods used to acquire the image. Compressive Sensing (CS) plays a major role in this, especially to unclog the Analog to Digital Converters which are generally representing the bottleneck of this type of sensors. In addition, CS eliminates traditional compression processing stages that are performed by embedded digital signal processors dedicated to this purpose. The interest is twofold because it allows both to consistently reduce the amount of data to be converted but also to suppress digital processing performed out of the sensor chip. For the moment, regarding the use of CS in image sensors, the main route of exploration as well as the intended applications aims at reducing power consumption related to these components (i.e. ADC & DSP represent 99% of the total power consumption). More broadly, the paradigm of CS allows to question or at least to extend the Nyquist-Shannon sampling theory. This thesis shows developments in the field of image sensors demonstrating that is possible to consider alternative applications linked to CS. Indeed, advances are presented in the fields of hyperspectral imaging, super-resolution, high dynamic range, high speed and non-uniform sampling. In particular, three research axes have been deepened, aiming to design proper architectures and acquisition processes with their associated reconstruction techniques taking advantage of image sparse representations. How the on-chip implementation of Compressed Sensing can relax sensor constraints, improving the acquisition characteristics (speed, dynamic range, power consumption) ? How CS can be combined with simple analysis to provide useful image features for high level applications (adding semantic information) and improve the reconstructed image quality at a certain compression ratio ? Finally, how CS can improve physical limitations (i.e. spectral sensitivity and pixel pitch) of imaging systems without a major impact neither on the sensing strategy nor on the optical elements involved ? A CMOS image sensor has been developed and manufactured during this Ph.D. to validate concepts such as the High Dynamic Range - CS. A new design approach was employed resulting in innovative solutions for pixels addressing and conversion to perform specific acquisition in a compressed mode. On the other hand, the principle of adaptive CS combined with the non-uniform sampling has been developed. Possible implementations of this type of acquisition are proposed. Finally, preliminary works are exhibited on the use of Liquid Crystal Devices to allow hyperspectral imaging combined with spatial super-resolution. The conclusion of this study can be summarized as follows: CS must now be considered as a toolbox for defining more easily compromises between the different characteristics of the sensors: integration time, converters speed, dynamic range, resolution and digital processing resources. However, if CS relaxes some material constraints at the sensor level, it is possible that the collected data are difficult to interpret and process at the decoder side, involving massive computational resources compared to so-called conventional techniques. The application field is wide, implying that for a targeted application, an accurate characterization of the constraints concerning both the sensor (encoder), but also the decoder need to be defined.

EPFL2015

Exploring information retrieval using image sparse representations

William Guicquéro

EPFL2015

Parallel architecture for high-speed analog-to-digital conversion

Guillaume Ding

EPFL2009

Processeur de signal numérique

Advanced signal processing algorithms for 3rd generation wireless mobile systems

Parallel architecture for high-speed analog-to-digital conversion

Exploring information retrieval using image sparse representations

Advanced signal processing algorithms for 3rd generation wireless mobile systems

Exploring information retrieval using image sparse representations

Parallel architecture for high-speed analog-to-digital conversion