Sustainably realizing applications of the future with high performance demands requires that energy efficiency becomes a central design criterion for the entire system. For example, the power consumption of the analog-to-digital converter (ADC) can become a major factor when transmitting at large bandwidths and carrier frequencies, e.g., for ultra-short range high data rate communication. The consumed energy per conversion step increases with the sampling rate such that high resolution ADCs become unfeasible in the sub-THz regime at the very high sampling rates required. This makes signaling schemes adapted to 1-bit quantizers a promising alternative. We therefore quantify the performance of bandlimited 1-bit quantized wireless communication channels using techniques like oversampling and faster-than-Nyquist (FTN) signaling to compensate for the loss of achievable rate. As a limiting case, we provide bounds on the mutual information rate of the hard bandlimited 1-bit quantized continuous-time – i.e., infinitely oversampled – additive white Gaussian noise channel in the mid-to-high signal-to-noise ratio (SNR) regime. We derive analytic expressions using runlength encoded input signals. For real signals the maximum value of the lower bound on the spectral efficiency in the high-SNR limit was found to be approximately 1.63 bit/s/Hz. Since in practical scenarios the oversampling ratio remains finite, we derive bounds on the achievable rate of the bandlimited oversampled discrete-time channel. These bounds match the results of the continuous-time channel remarkably well. We observe spectral efficiencies up to 1.53 bit/s/Hz in the high-SNR limit given hard bandlimitation. When excess bandwidth is tolerable, spectral efficiencies above 2 bit/s/Hz per domain are achievable w.r.t. the 95 %-power containment bandwidth. Applying the obtained bounds to a bandlimited oversampled 1-bit quantized multiple-input multiple-output channel, we show the benefits when using appropriate power allocation schemes. As a constant envelope modulation scheme, continuous phase modulation is considered in order to relieve linearity requirements on the power amplifier. Noise-free performance limits are investigated for phase shift keying (PSK) and continuous phase frequency shift keying (CPFSK) using higher-order modulation alphabets and intermediate frequencies. Adapted waveforms are designed that can be described as FTN-CPFSK. With the same spectral efficiency in the high-SNR limit as PSK and CPFSK, these waveforms provide a significantly improved bit error rate (BER) performance. The gain in SNR required for achieving a certain BER can be up to 20 dB. / Die nachhaltige Realisierung von zukünftigen Übertragungssystemen mit hohen Leistungsanforderungen erfordert, dass die Energieeffizienz zu einem zentralen Designkriterium für das gesamte System wird. Zum Beispiel kann die Leistungsaufnahme des Analog-Digital-Wandlers (ADC) zu einem wichtigen Faktor bei der Übertragung mit großen Bandbreiten und Trägerfrequenzen werden, z. B. für die Kommunikation mit hohen Datenraten über sehr kurze Entfernungen. Die verbrauchte Energie des ADCs steigt mit der Abtastrate, so dass hochauflösende ADCs im Sub-THz-Bereich bei den erforderlichen sehr hohen Abtastraten schwer einsetzbar sind. Dies macht Signalisierungsschemata, die an 1-Bit-Quantisierer angepasst sind, zu einer vielversprechenden Alternative. Wir quantifizieren daher die Leistungsfähigkeit von bandbegrenzten 1-Bit-quantisierten drahtlosen Kommunikationssystemen, wobei Techniken wie Oversampling und Faster-than-Nyquist (FTN) Signalisierung eingesetzt werden, um den durch Quantisierung verursachten Verlust der erreichbaren Rate auszugleichen. Wir geben Grenzen für die Transinformationsrate des Extremfalls eines strikt bandbegrenzten 1-Bit quantisierten zeitkontinuierlichen – d.h. unendlich überabgetasteten – Kanals mit additivem weißen Gauß’schen Rauschen bei mittlerem bis hohem Signal-Rausch-Verhältnis (SNR) an. Wir leiten analytische Ausdrücke basierend auf lauflängencodierten Eingangssignalen ab. Für reelle Signale ist der maximale Wert der unteren Grenze der spektralen Effizienz im Hoch-SNR-Bereich etwa 1,63 Bit/s/Hz. Da die Überabtastrate in praktischen Szenarien endlich bleibt, geben wir Grenzen für die erreichbare Rate eines bandbegrenzten, überabgetasteten zeitdiskreten Kanals an. Diese Grenzen stimmen mit den Ergebnissen des zeitkontinuierlichen Kanals bemerkenswert gut überein. Im Hoch-SNR-Bereich sind spektrale Effizienzen bis zu 1,53 Bit/s/Hz bei strikter Bandbegrenzung möglich. Wenn Energieanteile außerhalb des Frequenzbandes tolerierbar sind, können spektrale Effizienzen über 2 Bit/s/Hz pro Domäne – bezogen auf die Bandbreite, die 95 % der Energie enthält – erreichbar sein. Durch die Anwendung der erhaltenen Grenzen auf einen bandbegrenzten überabgetasteten 1-Bit quantisierten Multiple-Input Multiple-Output-Kanal zeigen wir Vorteile durch die Verwendung geeigneter Leistungsverteilungsschemata. Als Modulationsverfahren mit konstanter Hüllkurve betrachten wir kontinuierliche Phasenmodulation, um die Anforderungen an die Linearität des Leistungsverstärkers zu verringern. Beschränkungen für die erreichbare Datenrate bei rauschfreier Übertragung auf Zwischenfrequenzen mit Modulationsalphabeten höherer Ordnung werden für Phase-shift keying (PSK) and Continuous-phase frequency-shift keying (CPFSK) untersucht. Weiterhin werden angepasste Signalformen entworfen, die als FTN-CPFSK beschrieben werden können. Mit der gleichen spektralen Effizienz im Hoch-SNR-Bereich wie PSK und CPFSK bieten diese Signalformen eine deutlich verbesserte Bitfehlerrate (BER). Die Verringerung des erforderlichen SNRs zur Erreichung einer bestimmten BER kann bis zu 20 dB betragen.

[pt] Documentos digitalizados em ambientes de negócios substituíram grandes volumes de papéis. Profissionais autorizados usam carimbos para certificar informações críticas nesses documentos. Muitas empresas precisam verificar o carimbo adequado de documentos de entrada e saída. Na maioria das situações de inspeção, as pessoas realizam inspeção visual para identificar carimbos. Assim sendo, a verificação manual de carimbos é cansativa, suscetível a erros e ineficiente em termos de tempo gasto e resultados esperados. Erros na verificação manual de carimbos podem gerar multas de órgãos reguladores, interrupção de operações e até mesmo comprometer fluxos de trabalho e transações financeiras. Este trabalho propõe dois métodos que combinados podem resolver esse problema, automatizando totalmente a detecção de carimbos em documentos digitalizados do mundo real. Os métodos desenvolvidos podem lidar com conjuntos de dados contendo muitos tipos de carimbos de tamanho de amostra pequena, com múltiplas sobreposições, combinações diferentes por página e dados ausentes. O primeiro método propõe uma arquitetura de rede profunda projetada a partir da relação entre os problemas identificados em carimbos do mundo real e os desafios e soluções da tarefa de detecção de objetos apontados na literatura. O segundo método propõe um novo pipeline de aumento de instâncias de conjuntos de dados de carimbos a partir de dados reais e investiga se é possível detectar tipos de carimbos com amostras insuficientes. Este trabalho avalia os hiperparâmetros da abordagem de aumento de instâncias e os resultados obtidos usando um método Deep Explainability. Foram alcançados resultados de última geração para a tarefa de detecção de carimbos combinando com sucesso esses dois métodos, alcançando 97.3 por cento de precisão e 93.2 por cento de recall. / [en] Scanned documents in business environments have replaced large volumes of papers. Authorized professionals use stamps to certify critical information in these documents. Many companies need to verify the adequate stamping of incoming and outgoing documents. In most inspection situations, people perform a visual inspection to identify stamps. Therefore, manual stamp checking is tiring, susceptible to errors, and inefficient in terms of time spent and expected results. Errors in manual checking for stamps can lead to fines from regulatory bodies, interruption of operations, and even compromise workflows and financial transactions. This work proposes two methods that combined can address this problem, by fully automating stamp detection in real-world scanned documents. The developed methods can handle datasets containing many small sample-sized types of stamps, multiples overlaps, different combinations per page, and missing data. The first method proposes a deep network architecture designed from the relationship between the problems identified in real-world stamps and the challenges and solutions of the object detection task pointed out in the literature. The second method proposes a novel instance augmentation pipeline of stamp datasets from real data to investigate whether it is possible to detect stamp types with insufficient samples. We evaluate the hyperparameters of the instance augmentation approach and the obtained results through a Deep Explainability method. We achieve state-of-the-art results for the stamp detection task by successfully combining these two methods, achieving 97.3 percent of precision and 93.2 percent of recall.

Nowadays, several studies are launched for the design of reliable and safe communications systems that introduce Unmanned Aerial Vehicle (UAV), this paves the way for UAV communication systems to play an important role in a lot of applications for non-segregated military and civil airspaces. Until today, rules for integrating commercial UAVs in airspace still need to be defined, the design of secure, highly reliable and cost effective communications systems still a challenging task. This thesis is part of this communication context. Motivated by the rapid growth of UAV quantities and by the new generations of UAVs controlled by satellite, the thesis aims to study the various possible UAV links which connect UAV/aircraft to other communication system components (satellite, terrestrial networks, etc.). Three main links are considered: the Forward link, the Return link and the Mission link. Due to spectrum scarcity and higher concentration in aircraft density, spectral efficiency becomes a crucial parameter for largescale deployment of UAVs. In order to set up a spectrally efficient UAV communication system, a good understanding of transmission channel for each link is indispensable, as well as a judicious choice of the waveform. This thesis begins to study propagation channels for each link: a mutipath channels through radio Line-of-Sight (LOS) links, in a context of using Meduim Altitude Long drones Endurance (MALE) UAVs. The objective of this thesis is to maximize the solutions and the algorithms used for signal reception such as channel estimation and channel equalization. These algorithms will be used to estimate and to equalize the existing muti-path propagation channels. Furthermore, the proposed methods depend on the choosen waveform. Because of the presence of satellite link, in this thesis, we consider two low-papr linear waveforms: classical Single-Carrier (SC) waveform and Extented Weighted Single-Carrier Orthogonal Frequency-Division Multiplexing (EW-SC-OFDM) waveform. channel estimation and channel equalization are performed in the time-domain (SC) or in the frequency-domain (EW-SC-OFDM). UAV architecture envisages the implantation of two antennas placed at wings. These two antennas can be used to increase diversity gain (channel matrix gain). In order to reduce channel equalization complexity, the EWSC- OFDM waveform is proposed and studied in a muti-antennas context, also for the purpose of enhancing UAV endurance and also increasing spectral efficiency, a new modulation technique is considered: Spatial Modulation (SM). In SM, transmit antennas are activated in an alternating manner. The use of EW-SC-OFDM waveform combined to SM technique allows us to propose new modified structures which exploit exces bandwidth to improve antenna bit protection and thus enhancing system performances.

This work researches named entity recognition (NER) with respect to images of documents with a domain-specific layout, by means of Convolutional Neural Networks (CNNs). Examples of such documents are receipts, invoices, forms and scientific papers, the latter of which are used in this work. An NER task is first performed statically, where a static number of entity classes is extracted per document. Networks based on the deep VGG-16 network are used for this task. Here, experimental evaluation shows that framing the task as a classification task, where the network classifies each bounding box coordinate separately, leads to the best network performance. Also, a multi-headed architecture is introduced, where the network has an independent fully-connected classification head per entity. VGG-16 achieves better performance with the multi-headed architecture than with its default, single-headed architecture. Additionally, it is shown that transfer learning does not improve performance of these networks. Analysis suggests that the networks trained for the static NER task learn to recognise document templates, rather than the entities themselves, and therefore do not generalize well to new, unseen templates. For a dynamic NER task, where the type and number of entity classes vary per document, experimental evaluation shows that, on large entities in the document, the Faster R-CNN object detection framework achieves comparable performance to the networks trained on the static task. Analysis suggests that Faster R-CNN generalizes better to new templates than the networks trained for the static task, as Faster R-CNN is trained on local features rather than the full document template. Finally, analysis shows that Faster R-CNN performs poorly on small entities in the image and suggestions are made to improve its performance.

Different type of images provided by various combinations of cameras have the power to help increase and optimize plant growth. Along with a powerful deep learning model, for the purpose of detecting these stress indicators in RGB images, can significantly increase the harvest yield. The field of AI solutions in agriculture is not vastly explored and this thesis aims to take a first step in helping explore different techniques to detect early plant stress. Within this work, different types and combinations of camera modules will initially be reviewed and evaluated based on the amount of information they provide. Using the chosen cameras, we manually set up datasets and annotations, chose and then trained a suitable and appropriate algorithm to predict deviations from an ideal state in plant production. The algorithm chosen was Faster RCNN, which resulted in having a very high detection accuracy. Along with the main type of cameras, a new particular type of images analysis, named SI-NDVI, is done using a particular combination of the main three cameras and the results show that it is able to detect vegetation and able to predict or show if a plant is stressed or not. An in-depth research is done on all these techniques to create a good quality dataset for the purpose of early stress detection. / Olika typer av bilder försedda av olika kombinationer av kameror har kapaciteten att hjälpa öka och optimera odling av växter. Tillsammans med en kraftfull deep learning-modell, för att detektera olika stressindikatorer i RGB bilder, kan signifikant öka skördar. Fältet av AI-lösningar inom jordbruk är inte väl utforskat och denna uppsats siktar på att ta ett första steg i utforskandet av olika tekniker för att detektera tidig stress hos växter. Inom detta arbete kommer olika typer och kombinationer av kameramoduler bli utvärderade baserat på hur mycket information de kan förse. Genom att använda de valda kamerorna skapar vi själva dataseten och kategoriserar dem, därefter välja och träna en lämplig algoritm för att förutspå förändringar från ett idealt tillstånd för växtens tillväxt. Algoritmen som valdes var Faster RCNN, vilken hade en väldigt hög träffsäkerhet. Parallellt med de huvudsakliga kamerorna genomförs en ny typ av bildanalys vid namn SI-NDVI genom användandet av en särskild kombination av de tre kameror och resultat visar att det är möjligt att detektera vegetation och förutspå eller visa om en växt är stressad eller inte. En fördjupad undersökning genomförs på alla dessa tekniker för att skapa ett dataset av god kvalité för att kunna förutspå tidig stress.

The need for products has increased during the last few years, this high demand needs to bemet with higher means of production. The use of neural networks can be the key to increasedproduction without having to compromise product quality or human workers well being. This thesislooks into the concept of reliable architectures in convolutional neural networks and how they canbe implemented. The neural networks are trained to recognize the features in images to identifycertain objects, these recognition is then compared to other models to see which of them had the bestprediction. Using multiple models creates a reliable architecture from which results can be produced,these results can then be used in combinations with algorithms to improve prediction certainty. Theaim of implementing the networks with these algorithms are to improve the results without havingto change the networks configurations.

According to the "faster-X effect", X-linked genes and Z-linked genes evolve more quickly than autosomal genes. This theory is one of the currently intensively studied topics in evolutionary research. However, performing high quality tests is difficult because the results are influenced by several factors - the effective size of the population of the gonosome, sexual selection, the dependency of mutation rate and selection on sex, and the mechanism of dosage compensation. Conservation of genes and possible differences between rates of evolution of autosomes also play a role and not all studies take this fact into account. This study shows some of the difficulties of paired comparisons of dN/dS ratios traditionally used to test faster-X or faster-Z effects and introduces the basis of a new method of comparison of the rate of evolution (CREC) based on relative genetic distances between three species. The CREC method reduces the influence of conservation of genes on results and is more applicable for testing faster-X or faster Z effects in such species where two species without homologic gonosomes can be found for comparison. In means of the development of this method, the faster-Z effect was tested on a dataset of 9 autosomal and 13 Z-linked genes in the six-striped long-tailed lizard (Takydromus...

ITC/USA 2015 Conference Proceedings / The Fifty-First Annual International Telemetering Conference and Technical Exhibition / October 26-29, 2015 / Bally's Hotel & Convention Center, Las Vegas, NV / One of the requirements resulting from mounting pressure on flight test schedules is the reduction of time needed for data analysis, in pursuit of shorter test cycles. This requirement has ramifications such as the demand for record and processing of not just raw measurement data but also of data converted to engineering units in real time, as well as for an optimized use of the bandwidth available for telemetry downlink and ultimately for shortening the duration of procedures intended to disseminate pre-selected recorded data among different analysis groups on ground. A promising way to successfully address these needs consists in implementing more CPU-intelligence and processing power directly on the on-board flight test equipment. This provides the ability to process complex data in real time. For instance, data acquired at different hardware interfaces (which may be compliant with different standards) can be directly converted to more easy-to-handle engineering units. This leads to a faster extraction and analysis of the actual data contents of the on-board signals and busses. Another central goal is the efficient use of the available bandwidth for telemetry. Real-time data reduction via intelligent filtering is one approach to achieve this challenging objective. The data filtering process should be performed simultaneously on an all-data-capture recording and the user should be able to easily select the interesting data without building PCM formats on board nor to carry out decommutation on ground. This data selection should be as easy as possible for the user, and the on-board FTI devices should generate a seamless and transparent data transmission, making a quick data analysis viable. On-board data processing and filtering has the potential to become the future main path to handle the challenge of FTI data acquisition and analysis in a more comfortable and effective way.

In large scale productions of metal sheets, it is important to maintain an effective way to continuously inspect the products passing through the production line. The inspection mainly consists of detection of defects and tracking of ID numbers. This thesis investigates the possibilities to create an automatic inspection system by evaluating different machine learning algorithms for defect detection and optical character recognition (OCR) on metal sheet data. Digit recognition and defect detection are solved separately, where the former compares the object detection algorithm Faster R-CNN and the classical machine learning algorithm NCGF, and the latter is based on unsupervised learning using a convolutional autoencoder (CAE). The advantage of the feature extraction method is that it only needs a couple of samples to be able to classify new digits, which is desirable in this case due to the lack of training data. Faster R-CNN, on the other hand, needs much more training data to solve the same problem. NCGF does however fail to classify noisy images and images of metal sheets containing an alloy, while Faster R-CNN seems to be a more promising solution with a final mean average precision of 98.59%. The CAE approach for defect detection showed promising result. The algorithm learned how to only reconstruct images without defects, resulting in reconstruction errors whenever a defect appears. The errors are initially classified using a basic thresholding approach, resulting in a 98.9% accuracy. However, this classifier requires supervised learning, which is why the clustering algorithm Gaussian mixture model (GMM) is investigated as well. The result shows that it should be possible to use GMM, but that it requires a lot of GPU resources to use it in an end-to-end solution with a CAE.

This master's thesis deals with a vehicle detector based on the convolutional neural network and scene captured by drone. Dataset is described at the beginning, because the main aim of this thesis is to create practicly usable detector. Architectures of the forward neural networks which detector was created from are described in the next chapter. Techniques for building a detector based on the naive methods and current the most successful meta architectures follow the neural network architectures. An implementation of the detector is described in the second part of this thesis. The final detector was built on meta architecture Faster R-CNN and PVA neural network on which the detector achieved score over 90 % and 45 full HD frames per seconds.