Thesis advisor: Kevin Bedell / Using the AdS/CFT correspondence, it has been shown that the ratio of shear viscosity to entropy density is bounded from below in strongly coupled field theories with a gravity dual. More recently, this bound has been shown to be grossly violated in novel non-Fermi liquids and the unitary Fermi gas in the presence of superfluid fluctuations above T_c. Nevertheless, a holographic approach to such systems which break the lower bound have been strongly reliant on AdS spacetimes with massive gravitons. In this work, we propose a violation of the viscosity over entropy bound in 3+1 dimensional AdS spacetimes that support stable black hole solutions with non-zero scalar field. Such a black hole is shown to be characterized by a novel phase transition at large negative mass, where the underlying thermodynamics agrees with the Larkin-Ovchinnikov-Fulde-Ferrell (LOFF)-like phase seen in the unitary Fermi gas near Tc and the bound is similarly broken. Such a work paves the way for a holographic description of strongly-entangled quantum fluids at high Reynolds number. / Thesis (BS) — Boston College, 2021. / Submitted to: Boston College. College of Arts and Sciences. / Discipline: Scholar of the College. / Discipline: Physics.

Driver sleepiness contributes to a large amount of all road traffic crashes. Developing an objective measurement of driver sleepiness in order to prevent eventual traffic accidents is desirable. The aim of this master thesis was to investigate if deep learning can be used to provide a driver sleepiness classification from brain activity signals obtained by electroencephalography (EEG). The intention was to study the classification performance when using different representations of the input data and to examine how various deep neural network architectures and class weighting during training affect the classification.  The data was collected from 12 experiments, where 269 participants (1187 driving sessions) were driving either on real roads or in a moving-base driving simulator, while electrophysiological data was recorded. Several deep neural network architectures were developed, depending on the representation of the input data.  Regardless of which data representation that was used as input to the network, the datawas divided into three datasets: Training 60%, validation 20% and test 20%. The data from each participant, with associated driving sessions, were randomly assigned to the different datasets according to the given percentage, which resulted in a subject-independent sleepiness detection. The output was in the form of continuous regression further rounded to the closest integer and divided into five classes according to Karolinska Sleepiness Scale (KSS = 1-5, 6, 7, 8, 9). The best performance was obtained with a convolutional neural network (CNN) combined with Long Short-Term Memory (LSTM) architecture, with time series data as input. This gave an accuracy of 41.44%, a mean absolute error of 0.94 and a macro F1-score of 0.37. Overall, the models with time series data showed better classification results compared to those with time-frequency data. Class weighting, giving all classes inverse proportional weight to their appearance, compensated slightly for class imbalance, but all networks had in general difficulties with generalizing to new data.

Driver sleepiness is a cause for crashes and it is estimated that 3.9 to 33 % of all crashes might be related to sleepiness at the wheel. It is desirable to get an objective measurement of driver sleepiness for reduced sensitivity to subjective variations. Using deep learning for classification of driver sleepiness could be a step toward this objective. In this master thesis, deep learning was used for investigating classification of electrophysiological data, electroencephalogram (EEG) and electrooculogram (EOG), from drivers into levels of sleepiness. The EOG reflects eye position and EEG reflects brain activity. Initially, the intention was to include electrocardiogram (ECG), which reflects heart activity, in the research but this data were later excluded. Both raw time series data and data transformed into time-frequency domain representations were fed into the developed neural networks and for comparison manually extracted features were used in a shallow neural network architecture. Investigation of using EOG and EEG data separately as input was performed as well as a combination as input. The data were labeled using the Karolinska Sleepiness Scale, and the scale was divided into two labels "fatigue" and "alert" for binary classification or in five labels for comparison of classification and regression. The effect of example length was investigated using 150 seconds, 60 seconds and 30 seconds data. Different variations of the main network architecture were used depending on the data representation and the best result was given when using a combination of a convolutional neural network (CNN) and a long short-term memory (LSTM) network with time distributed 150 seconds EOG data as input. The accuracy was in this case 80.4 % and the majority of both alert and fatigue epochs were classified correctly with 85.7 % and 66.7 % respectively. Using the optimal threshold from the created receiver operating characteristics (ROC) curve resulted in a more balanced classifier with 76.3 % correctly classified alert examples and 79.2 % correctly classified fatigue examples. The results from the EEG data, both in terms of accuracy and distribution of correctly classified examples, were shown to be less promising compared to EOG data. Combining EOG and EEG signals was shown to slightly increase the proportion of correctly classified fatigue examples. However, more promising results were obtained when balancing the classifier for solely EOG signals. The overall result from this project shows that there are patterns in the data connected to sleepiness that the neural network can find which makes further work on applying deep learning to the area of driver sleepiness interesting.

The three years lasting postwar period in Czechoslovak history, sometimes denoted as the "Third Republic", still deserves attention of the historical research in the Czech Republic. The 1947 autumn political crisis in Slovakia is undoubtly one of the most important issues of that historical period. The aim of this thesis is to point out the sources, the process and consequences of this political crisis that is nowadays recognized as a "litmus paper" of the following "coup d'etat". First parts of the thesis analyse the collapse of the nondemocratic regime of the Slovak Republic as well as the framework of the political system of Czechoslovakia and its main features. The following part deals with circumstances of the general elections that took place in May 1946, elections, that resulted in a quite unexpected win of the Democratic Party in the Slovak part of Czechoslovakia. The political situation in Czechoslovakia deteriorated, as the year 1947 continued. It developed into a deep political crisis in autumn, preceded by a campaign against the alleged subversive complot against the state authority. The complot and the following crisis form the main theme of the thesis. These occasions deserve maximal attention because they explicitly show that the decisive politics was dominated by the bloc...

Falling asleep while operating a moving vehicle is a contributing factor to the statistics of road related accidents. It has been estimated that 20% of all accidents where a vehicle has been involved are due to sleepiness behind the wheel. To prevent accidents and to save lives are of uttermost importance. In this thesis, given the world’s largest dataset of driver participants, two methods of evaluating driver sleepiness have been evaluated. The first method was based on the creation of epochs from lane departures and KSS, whilst the second method was based solely on the creation of epochs based on KSS. From the epochs, a number of features were extracted from both physiological signals and the car’s controller area network. The most important features were selected via a feature selection step, using sequential forward floating selection. The selected features were trained and evaluated on linear SVM, Gaussian SVM, KNN, random forest and adaboost. The random forest classifier was chosen in all cases when classifying previously unseen data.The results shows that method 1 was prone to overfit. Method 2 proved to be considerably better, and did not suffer from overfitting. The test results regarding method 2 were as follows; sensitivity = 80.3%, specificity = 96.3% and accuracy = 93.5%.The most prominent features overall were found in the EEG and EOG domain together with the sleep/wake predictor feature. However indications have been made that complexities might contribute to the detection of sleepiness as well, especially the Higuchi’s fractal dimension.

Sleep is a basic need as important as physical fitness and good nutrition. Without enough sleep, we will create a sleep debt and experience sleepiness. Sleepiness can be defined as the inability to stay awake, a condition that has become a health problem in our 24-hour-7-day-a-week society. Estimates suggest that up to one-third of the population suffers from excessive sleepiness. Among other interactions, sleepiness affects our performance, increasing the risk of being involved in accidents. A considerable portion of work related accidents and injuries are related to sleepiness resulting in large costs for the individuals and society. Professional drivers are one example of workers who are at risk of sleepiness related accidents. Up to 40% of heavy truck accidents could be related to sleepiness. A better knowledge about reliable indicators and predictors of sleepiness is important in preventing sleepiness related accidents. This thesis investigates both objective and subjective indicators of sleepiness, how these relate to each other, and how their pattern changes over time. The indicators investigated were electroencephalography, heart rate variability, simple reaction time, head movement, and subjective ratings of sleepiness (Study I-IV). In Study V, a questionnaire study was conducted with professional drivers in northern Sweden. This study mainly deals with predictors of sleepiness. When subjects were sleep deprived both objective and subjective ratings indicated a rapid increase in sleepiness during the first hour of the test followed by a levelling off. This change in pattern was evident for all the indicators except heart rate and heart rate variability. On the other hand, HRV was correlated with the increase of EEG parameters during the post-test sleep period. The changes in pattern of the indicators included in the thesis are analysed in the perspective of temporal patterns and relationships. Of the tested indicators, a subjective rating of sleepiness with CR-10 was considered to be the most reliable indicator of sleepiness. Of the investigated predictors of sleepiness, prior sleep habits were found to be strongly associated to sleepiness and the sleepiness related symptoms while driving. The influences of driving conditions and individual characteristics on sleepiness while driving were lower. A multidisciplinary approach when investigating and implementing indicators and predictors of sleepiness is important. In addition to their actual relations to the development of sleepiness, factors such as technical and practical limitations, work, and individual and situational needs must be taken into account.

The research conducted for this thesis seeks to understand issues associated with integrating a direct spread spectrum system (DSSS) transceiver on to a single chip. Various types of sequences, such as Kasami sequences and Gold sequences, are available for use in typical spread spectrum systems. For this thesis, complex spreading sequences (CSS) are used for improved cross-correlation and autocorrelation properties that can be achieved by using such a sequence. While CSS and DSSS are well represented in the existing body of knowledge, and discrete bulky hardware solutions exist – an effort to jointly integrate CSS and DSSS on-chip was identified to be lacking. For this thesis, spread spectrum architecture was implemented focussing on sub-systems that are specific to CSS. This will be the main contribution for this thesis, but the contribution is further appended by various RF design challenges: highspeed requirements make RF circuits sensitive to the effects of parasitics, including parasitic inductance, passive component modelling, as well as signal integrity issues. The integration is first considered more ideally, using mathematical sub-systems, and then later implemented practically using complementary metal-oxide semiconductor (CMOS) technology. The integration involves mixed-signal and radio frequency (RF) design techniques – and final integration involves several specialized analogue sub-systems, such as a class F power amplifier (PA), a low-noise amplifier (LNA), and LC voltage-controlled oscillators (VCOs). The research also considers various issues related to on-chip inductors, and also considers an active inductor implementation as an option for the VCO. With such an inductor a better quality factor is achievable. While some conventional sub-system design techniques are deployed, several modifications are made to adapt a given sub-system to the design requirements for this thesis. The contribution of the research lies in the circuit level modifications done at sub-system level aimed towards eventual integration. For multiple-access communication systems, where a number of independent users are required to share a common channel, the transceiver proposed in this thesis, can contribute towards improved data rate or bit error rate. The design is completed for fabrication in a standard 0.35-μm CMOS process with minimal external components. With an active chip area of about 5 mm2, the simulated transmitter consumes about 250 mW&the receiver consumes about 200 mW. AFRIKAANS : Die navorsing wat vir hierdie tesis onderneem is, beoog om kundigheid op te bou aangaande die kwessies wat met die integrasie van ‘n direkte spreispektrumstelsel (DSSS) sender-ontvanger op ‘n enkele skyfie verband hou. Verskeie tipes sekwensies, soos byvoorbeeld Kasami- en Gold-sekwensies, is vir gebruik in tipiese spreispektrumstelsels beskikbaar. Vir hierdie tesis is komplekse spreisekwensies (KSS) gebruik vir verbeterde kruis- en outokorrelasie-eienskappe wat bereik kan word deur so ‘n sekwensie te gebruik. Alhoewel DSSS en KSS reeds welbekend is, en diskrete hardeware oplossings reeds bestaan, is die vraag na gesamentlike geïntegreerde DSSS en KSS op een vlokkie geïdentifiseer. Vir hierdie tesis is spreispektrumargitektuur aangewend met die klem op KSS substelsels. Dit is dan ook die belangrikste bydrae van hierdie tesis, maar die bydrae gaan verder gepaard met verskeie RF-ontwerpuitdagings: hoëspoed-vereistes maak RF-stroombane sensitief vir die uitwerking van parasitiese komponente, met inbegrip van parasitiese induktansie, passiewe komponentmodellering en ook seinintegriteitskwessies. Die integrasie word eerstens meer idealisties oorweeg deur wiskundige substelsels te gebruik en dan later prakties te implementeer deur komplementêre metaaloksied-halfgeleiertegnologie (CMOS) te gebruik. Die integrasie behels gemengdesein- en radiofrekwensie(RF)-ontwerptegnieke – en finale integrasie behels verskeie gespesialiseerde analoë substelsels soos ‘n klas F-kragversterker (KV), ‘n laeruis-versterker (LRV), en LC-spanningbeheerde ossileerders (SBO’s). Die navorsing oorweeg ook verskeie kwessies in verband met op-skyfie induktors en oorweeg ook ‘n aktiewe induktorimplementering as ‘n opsie vir die SBO. Met sodanige induktor is ‘n beter kwaliteitsfaktor haalbaar. Hoewel enkele konvensionele substelsel-ontwerptegnieke aangewend word, word daar verskeie wysigings aangebring om ‘n gegewe substelsel by die ontwerpvereistes vir hierdie tesis aan te pas. Die bydrae van die navorsing is hoofsaaklik die stroombaanmodifikasies wat gedoen is op substelselvlak om integrasie te vergemaklik. Vir veelvoudige-toegang kommunikasiestelsels waar ‘n aantal onafhanklike gebruikers dieselfde seinkanaal moet deel, kan die sender-ontvanger voorgestel in hierdie tesis meewerk om die datatempo en fouttempo te verbeter. Die ontwerp is voltooi vir vervaardiging in ‘n standaard 0.35-μm CMOS-proses met minimale eksterne komponente. Met ‘n aktiewe skyfie-oppervlakte van ongeveer 5 mm2, verbruik die gesimuleerde sender ongeveer 250 mW en die ontvanger verbruik ongeveer 200 mW. / Thesis (PHD)--University of Pretoria, 2011. / Electrical, Electronic and Computer Engineering / unrestricted

Bibliographic record: Michal MACHÁČEK: Gustáv Husák. A Political Biography with Special Emphasis on Czech-Slovak Relations in 20th Century, PhD. thesis, The Faculty of Arts at Charles University 2017, 476pp. [784 standard pages]. Abstract The dissertation thesis discusses public activities, thoughts and the political life of JUDr. Gustáv Husák, CSc. (1913-1991), who was involved in the Czech-Slovak public space for sixty years with a significant footprint even today. The text is based on a thorough research and is chronologically structured, intertwined with thematic areas, however an analytical approach prevails. The first chapter focuses on Husák's youth, the factors that led him to the communist movement, and his early activism. This is followed by a portrayal of the Husák's activities during the Second World War, his role in the resistance, participation in a propaganda trip to the Nazi conquered Ukraine, and his vision of Slovakia as a republic of the Soviet Union. His later involvement in the Slovak National Uprising provided the legitimacy of his later political career in the post-war era, when he successfully led the struggle for the communist monopoly of political power in Czechoslovakia and attempts to present the Communist Party of Slovakia as a national party. Next two chapters show the origins...