The aim of this thesis was to develop user interface in LabVIEW to make typical measurements with spectral analyzer HP89410A There is introduced the theory of operation of an analogue heterodyne and a digital FFT spectral analyzer. It is explained the background of the Fast Fourier Transform. There are discussed the key settings in measuring with FTT analyzer, like window function, bandwidth, number of frequency points, or the averaging. Further is described the program, which was developed for measuring frequency characteristic by white noise and by stepped measurement. It was also made a module for measuring Signal to Noise Ratio and module for Total Harmonic Distortion. In the last part of this thesis are shown the results of processed exemplar measurements.

This work deals with frequency generators based on the direct digital synthesis method DDS. Basic principles and attributes of the frequency generator DDS are explained. The text describes parameters influencing and defining a quality of the generated signal. The list of available integrated circuits realizing the direct digital synthesis is mentioned. A construction of the DDS generator with a device AD9954 and the generator control are described. At the end of the work, the function and parameters of the designed generator are verified.

Secret key generation from reciprocal multi-antenna channels is an interesting alternative to cryptographic key management in wireless systems without infrastructure access. In this work, we study the secret key rate for the basic source model with a MIMO channel. First, we derive an expression for the secret key rate under spatial correlation modelled by the Kronecker model and with spatial precoding at both communication nodes. Next, we analyze the result for uncorrelated antennas to understand the optimal precoding for this special case, which is equal power allocation. Then, the impact of correlation is characterized using Majorization theory. Surprisingly for small SNR, spatial correlation increases the secret key rate. For high SNR, the maximum secret key rate is achieved for uncorrelated antennas. The results indicate that a solid system design for reciprocal MIMO key generation is required to establish the secret key rate gains.

1-bit analog-to-digital conversion is very attractive for low-complexity communications receivers. A major drawback is, however, the small spectral efficiency when sampling at symbol rate. This can be improved through oversampling by exploiting the signal distortion caused by the transmission channel. This paper analyzes the achievable data rate of band-limited communications channels that are subject to additive noise and inter-symbol-interference with 1-bit quantization and oversampling at the receiver. It is shown that not only the channel noise but also the inter-symbol-interference can be exploited to benefit from oversampling.

In this paper, we analyze the achievable secrecy rates in the two-hop wiretap channel with four nodes, where the transmitter and the receiver have multiple antennas while the relay and the eavesdropper have only a single antenna each. The relay is operating in amplify-and-forward mode and all the channels between the nodes are known perfectly by the transmitter. We discuss different transmission and protection schemes like artificial noise (AN). Furthermore, we introduce interference neutralization (IN) as a new protection scheme. We compare the different schemes regarding the high-SNR slope and the high-SNR power offset and illustrate the performance by simulation results. It is shown analytically as well as by numerical simulations that the high SNR performance of the proposed IN scheme is better than the one of AN.

Blind source separation (BSS) aims at separating useful signal content from distortions. In the contactless acquisition of vital signs by means of the camera-based photoplethysmogram (cbPPG), BSS has evolved the most widely used approach to extract the cardiac pulse. Despite its frequent application, there is no consensus about the optimal usage of BSS and its general benefit. This contribution investigates the performance of BSS to enhance the cardiac pulse from cbPPGs in dependency to varying input data characteristics. The BSS input conditions are controlled by an automated spatial preselection routine of regions of interest. Input data of different characteristics (wavelength, dominant frequency, and signal quality) from 18 postoperative cardiovascular patients are processed with standard BSS techniques, namely principal component analysis (PCA) and independent component analysis (ICA). The effect of BSS is assessed by the spectral signal-tonoise ratio (SNR) of the cardiac pulse. The preselection of cbPPGs, appears beneficial providing higher SNR compared to standard cbPPGs. Both, PCA and ICA yielded better outcomes by using monochrome inputs (green wavelength) instead of inputs of different wavelengths. PCA outperforms ICA for more homogeneous input signals. Moreover, for high input SNR, the application of ICA using standard contrast is likely to decrease the SNR.

Recently, Deep Learning has revolutionized many fields, where one such area is Voice Activity Detection (VAD). This is of great interest to sectors of society concerned with detecting speech in sound signals. One such sector is the police, where criminal investigations regularly involve analysis of audio material. Convolutional Neural Networks (CNN) have recently become the state-of-the-art method of detecting speech in audio. But so far, understanding the impact of noise and sampling rates on such methods remains incomplete. Additionally, there are evaluation metrics from neighboring fields that remain unintegrated into VAD. We trained on four different sampling rates and found that changing the sampling rate could have dramatic effects on the results. As such, we recommend explicitly evaluating CNN-based VAD systems on pertinent sampling rates. Further, with increasing amounts of white Gaussian noise, we observed better performance by increasing the capacity of our Gated Recurrent Unit (GRU). Finally, we discuss how careful consideration is necessary when choosing a main evaluation metric, leading us to recommend Polyphonic Sound Detection Score (PSDS).

[ES] Esta tesis presenta diversas propuestas encaminadas a mejorar la transmisión de vídeo a través del estándar DASH (Dynamic Adaptive Streaming over HTTP). Este trabajo de investigación estudia el protocolo de transmisión DASH y sus características. A la vez, plantea la codificación con calidad constante y bitrate variable como modo de codificación del contenido de vídeo más indicado para la transmisión de contenido bajo demanda mediante el estándar DASH. Derivado de la propuesta de utilización del modo de codificación de calidad constante, cobra mayor importancia el papel que juegan los algoritmos de adaptación en la experiencia de los usuarios al consumir el contenido multimedia. En este sentido, esta tesis presenta un algoritmo de adaptación denominado Look Ahead el cual, sin modificar el estándar, permite utilizar la información de los tamaños de los segmentos de vídeo incluida en los contenedores multimedia para evitar tomar decisiones de adaptación que desemboquen en paradas no deseadas en la reproducción de contenido multimedia. Con el objetivo de evaluar las posibles mejoras del algoritmo de adaptación presentado, se proponen tres modelos de evaluación objetiva de la QoE. Los modelos propuestos permiten predecir de forma sencilla la QoE que tendrían los usuarios de forma objetiva, utilizando parámetros conocidos como el bitrate medio, el PSNR (Peak Signal-to-Noise Ratio) y el valor de VMAF (Video Multimethod Assessment Fusion). Todos ellos aplicados a cada segmento. Finalmente, se estudia el comportamiento de DASH en entornos Wi-Fi con alta densidad de usuarios. En este contexto, se producen un número elevado de paradas en la reproducción por una mala estimación de la tasa de transferencia disponible debida al patrón ON/OFF de descarga de DASH y a la variabilidad del acceso al medio de Wi-Fi. Para paliar esta situación, se propone un servicio de coordinación basado en la tecnología SAND (MPEG's Server and Network Assisted DASH) que proporciona una estimación de la tasa de transferencia basada en la información del estado de los players de los clientes. / [CA] Aquesta tesi presenta diverses propostes encaminades a millorar la transmissió de vídeo a través de l'estàndard DASH (Dynamic Adaptive Streaming over HTTP). Aquest treball de recerca estudia el protocol de transmissió DASH i les seves característiques. Alhora, planteja la codificació amb qualitat constant i bitrate variable com a manera de codificació del contingut de vídeo més indicada per a la transmissió de contingut sota demanda mitjançant l'estàndard DASH. Derivat de la proposta d'utilització de la manera de codificació de qualitat constant, cobra major importància el paper que juguen els algorismes d'adaptació en l'experiència dels usuaris en consumir el contingut. En aquest sentit, aquesta tesi presenta un algoritme d'adaptació denominat Look Ahead el qual, sense modificar l'estàndard, permet utilitzar la informació de les grandàries dels segments de vídeo inclosa en els contenidors multimèdia per a evitar prendre decisions d'adaptació que desemboquin en una parada indesitjada en la reproducció de contingut multimèdia. Amb l'objectiu d'avaluar les possibles millores de l'algoritme d'adaptació presentat, es proposen tres models d'avaluació objectiva de la QoE. Els models proposats permeten predir de manera senzilla la QoE que tindrien els usuaris de manera objectiva, utilitzant paràmetres coneguts com el bitrate mitjà, el PSNR (Peak Signal-to-Noise Ratio) i el valor de VMAF (Video Multimethod Assessment Fusion). Tots ells aplicats a cada segment. Finalment, s'estudia el comportament de DASH en entorns Wi-Fi amb alta densitat d'usuaris. En aquest context es produeixen un nombre elevat de parades en la reproducció per una mala estimació de la taxa de transferència disponible deguda al patró ON/OFF de descàrrega de DASH i a la variabilitat de l'accés al mitjà de Wi-Fi. Per a pal·liar aquesta situació, es proposa un servei de coordinació basat en la tecnologia SAND (MPEG's Server and Network Assisted DASH) que proporciona una estimació de la taxa de transferència basada en la informació de l'estat dels players dels clients. / [EN] This thesis presents several proposals aimed at improving video transmission through the DASH (Dynamic Adaptive Streaming over HTTP) standard. This research work studies the DASH transmission protocol and its characteristics. At the same time, this work proposes the use of encoding with constant quality and variable bitrate as the most suitable video content encoding mode for on-demand content transmission through the DASH standard. Based on the proposal to use the constant quality encoding mode, the role played by adaptation algorithms in the user experience when consuming multimedia content becomes more important. In this sense, this thesis presents an adaptation algorithm called Look Ahead which, without modifying the standard, allows the use of the information on the sizes of the video segments included in the multimedia containers to avoid making adaptation decisions that lead to undesirable stalls during the playback of multimedia content. In order to evaluate the improvements of the presented adaptation algorithm, three models of objective QoE evaluation are proposed. These models allow to predict in a simple way the QoE that users would have in an objective way, using well-known parameters such as the average bitrate, the PSNR (Peak Signal-to-Noise Ratio) and the VMAF (Video Multimethod Assessment Fusion). All of them applied to each segment. Finally, the DASH behavior in Wi-Fi environments with high user density is analyzed. In this context, there could be a high number of stalls in the playback because of a bad estimation of the available transfer rate due to the ON/OFF pattern of DASH download and to the variability of the access to the Wi-Fi environment. To relieve this situation, a coordination service based on SAND (MPEG's Server and Network Assisted DASH) is proposed, which provides an estimation of the transfer rate based on the information of the state of the clients' players. / Belda Ortega, R. (2021). Mejora del streaming de vídeo en DASH con codificación de bitrate variable mediante el algoritmo Look Ahead y mecanismos de coordinación para la reproducción, y propuesta de nuevas métricas para la evaluación de la QoE [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/169467

This thesis addresses the challenge of denoising microscopy images captured under low-light conditionswith varying intensity levels. The study compares three deep learning models — N2V, CARE, andRCAN — against the collaborative filter BM4D, which serves as a reference point. The models weretrained on two distinct datasets: Endoplasmic Reticulum and Mitochondria datasets, both acquired witha lattice light-sheet microscope.Results show that BM4D maintains stable performance metrics and delivers superior visual quality,when compared to the noisy input. In contrast, the deep learning models exhibit poor performance onnoisy test images when trained on datasets with non-uniform noise levels. Additionally, a sensitivitycomparison of neural parameter between the same models was made. Revealing that supervised modelsare data-specific to some extent, whereas the self-supervised N2V demonstrates consistent neuralparameters, suggesting lower data specificity. / Denna uppsats tar upp problemet med att reducera brus i mikroskopibilder tagna under svagaljusförhållanden med varierande intensitetsnivåer. Studien jämför tre djupinlärningsmodeller – N2V,CARE och RCAN – mot det kollaborativa filtret BM4D, vilket agerar som en referenspunkt.Modellerna tränades på två olika dataset: Endoplasmic Reticulum och Mitochondria, båda tagna meden selektiv planbelysningsmikroskop (lattice light-sheet microscope).Resultaten visar att BM4D behåller stabila prestationsmått och levererar bättre visuell kvalitet, jämförtmed den brusiga input. Däremot visar djupinlärningsmodellerna bristande prestanda på brusigatestbilder när de tränats på data med icke-enhetliga brusnivåer. Dessutom gjordes enkänslighetsjämförelse av neurala parametrar mellan samma modeller. Detta visade att de övervakademodellerna är specifika för data i viss utsträckning, medan den självövervakade N2V-modellen visarlika neurala parametrar, vilket tyder på lägre dataspecificitet

Imager des objets non-périodiques à une échelle nanométrique et à une échelle femto seconde est un vrai challenge à notre époque. Les techniques d’imagerie « sans lentille » sont des moyens puissants pour répondre à ce besoin. En utilisant des sources ultrarapide (~fs) et cohérente (ex. laser à électron libre ou harmoniques d’ordres élevés), ces techniques nous permettent de reconstruire des objets à partir de leur figure de diffraction, remplaçant les optiques conventionnelles du système d’imagerie par un algorithme informatique. Dans ce travail de thèse, je présent des expériences d’imageries en utilisant un rayonnement extrême-UV (15~40 nm) produit par la génération d’harmoniques d’ordre élevé d’un laser infrarouge puissant. Ce manuscrit est constitué d’une introduction, un chapitre de background théorique, trois chapitres de travail de thèse et une conclusion générale avec perspectives. La première partie du travail de thèse porte sur les développements et caractérisations de la ligne de lumière avec l’objectif de générer maximum de photons harmoniques cohérents avec un front d’onde plat. La deuxième partie est consacrée aux expériences et analyses de trois techniques d’imageries « sans lentille » : Imagerie par diffraction cohérente (CDI), Holographie par la transformée de Fourier (FTH) et Holographie avec références étendues (HERALDO). Ces derniers nous permettent de reconstruire des objets avec une résolution spatiale de 78 nm dans le cas de CDI et de 112 nm dans le cas de HERALDO, tous les deux avec une résolution temporaire de 20 fs. La troisième partie est une première application physique de l’imagerie sur la ligne harmonique. Il s’agit des études statiques et dynamiques de nano-domaines magnétique avec une résolution spatiale sub-100 nm à l’échelle femto seconde. Perspective des techniques d’imagerie 3D et développement potentiel de la ligne d’harmoniques sont présentés à la fin. / Ultrafast imaging of isolated objects with nanometric spatial resolution is a great challenge in our time. The lensless imaging techniques have shown great potential to answer this challenge. In lensless imaging, one can reconstruct sample images from their diffraction patterns with computational algorithms, which replace the conventional lens systems. Using ultrafast and coherent light sources, such as free electron laser and high order harmonics, one can investigate dynamic phenomena at the femtosecond time scale. In this thesis work, I present the lenless imaging experiments using XUV radiation provided by a laser driven high order harmonic beamline. The manuscript is composed of an introduction, a chapter of theoretical background, three chapters of main research work and a general conclusion with perspectives. The first part of this work concerns the development of the harmonic beamline to optimize the illumination condition for lensless imaging. The second part concentrates on the imaging techniques: the Coherent Diffraction Imaging (CDI), the Fourier Transform Holography (FTH) and the Holography using extended references (HERALDO). The reconstructions have achieved 78 nm spatial resolution in case of CDI and 112 nm resolution in case of HERALDO, both in single-shot regime corresponding to a temporal resolution of 20 fs. The third part presents the first physical application on the harmonic beamline using the lensless imaging. Samples with magnetic nano-domains have been studied with sub-100 nm spatial resolution, which paves the way for ultrafast magnetic dynamic studies. At the end, single-shot 3D imaging and further beamline development have been discussed.