As Moore's law nears its limit, a new form of signal processing is needed. Neuromorphic computing has used inspiration from biology to produce a new form of signal processing by mimicking biological neural networks using electrical components. Neuromorphic computing requires less signal preprocessing than digital systems since it can encode signals directly using analog temporal encoders from Spiking Neural Networks (SNNs). These encoders receive an analog signal as an input and generate a spike or spike trains as their output. The proposed temporal encoders use latency and Inter-Spike Interval (ISI) encoding and are expected to produce a highly sensitive hardware implementation of time encoding to preprocess signals for dynamic neural processors. Two ISI and two latency encoders were designed using Integrate-and-Fire (IF) and Leaky Integrate-and-Fire (LIF) neurons and optimized to produce low area designs. The IF and LIF neurons were designed using the Global Foundries 180nm CMOS process and achieved an area of 186µm2 and 182µm2, respectively. All four encoders have a sampling frequency of 50kHz. The latency encoders achieved an average energy consumption per spike of 277nJ and 316pJ for the IF-based and LIF-based latency encoders, respectively. The ISI encoders achieved an average energy consumption per spike of 1.07uJ and 901nJ for the IF-based and LIF-based ISI encoders, respectively. Power consumption is proportional to the number of neurons employed in the encoder and the potential to reduce power consumption through layout-level simulations is presented. The LIF neuron is able to use a smaller membrane capacitance to achieve similar operability as the IF neuron and consumes less area despite having more components. This demonstrates that capacitor sizes are the main limitations of a small size in spiking neurons for SNNs. An overview of the design and layout process of the two presented neurons is discussed with tips for overcoming problems encountered. The proposed designs can result in a fast neuromorphic process by employing a frequency higher than 10kHz and by providing a hardware implementation that is efficient in multiple sectors like machine learning, medical implementations, or security systems since hardware is safer from hacks. / Master of Science / As Moore's law nears its limit, a new form of signal processing is needed. Moore's law anticipated that transistor sizes will decrease exponentially as the years pass but CMOS technology is reaching physical limitations which could mean an end to Moore's prediction. Neuromorphic computing has used inspiration from biology to produce a new form of signal processing by mimicking biological neural networks using electrical components. Biological neural networks communicate through interconnected neurons that transmit signals through synapses. Neuromorphic computing uses a subdivision of Artificial Neural Networks (ANNs) called Spiking Neural Networks (SNNs) to encode input signals into voltage spikes to mimic biological neurons. Neuromorphic computing reduces the preprocessing step needed to process data in the digital domain since it can encode signals directly using analog temporal encoders from SNNs. These encoders receive an analog signal as an input and generate a spike or spike trains as their output. The proposed temporal encoders use latency and Inter-Spike Interval (ISI) encoding and are expected to produce a highly sensitive hardware implementation of time encoding to preprocess signals for dynamic neural processors. Two ISI and two latency encoders were designed using Integrate-and-Fire (IF) and Leaky Integrate-and-Fire (LIF) neurons and optimized to produce low area designs. All four encoders have a sampling frequency of 50kHz. The latency encoders achieved an average energy consumption per spike of 277nJ and 316pJ for the IF-based and LIF-based latency encoders, respectively. The ISI encoders achieved an average energy consumption per spike of 1.07uJ and 901nJ for the IF-based and LIF-based ISI encoders, respectively. Power consumption is proportional to the number of neurons employed in the encoder and the potential to reduce power consumption through layout-level simulations is presented. The LIF neuron is able to use a smaller membrane capacitance to achieve similar operability which consumes less area despite having more components than the IF neuron. This demonstrates that capacitor sizes are the main limitations of small size in neurons for spiking neural networks. An overview of the design and layout process of the two presented neurons is discussed with tips for overcoming problems encountered. The proposed designs can result in a fast neuromorphic process by employing a frequency higher than 10kHz and by providing a hardware implementation that is efficient in multiple sectors like machine learning, medical implementations, or security systems since hardware is safer from hacks.

This thesis explores a method for identifying JPEG encoders without relying on metadata by analyzing characteristics inherent to the JPEG file format itself. The approach uses machine learning to differentiate encoders based on features such as quantization tables, Huffman tables, and marker sequences. These features are extracted from the file container and analyzed to identify the source encoder. The random forest classification algorithm was applied to test the efficacy of the approach across different datasets, aiming to validate the model's performance and reliability. The results confirm the model's capability to identify JPEG source encoders, providing a useful approach for digital forensic investigations.

<p>A power assisted wheelchair combines human power, which is delivered by the arms through the pushrims, with electrical motors, which are powered by a battery. Today’s electric power assisted wheelchairs use force sensors to measure the torque exerted on the pushrims by the user. The force sensors in the pushrims are rather expensive and this approach also makes the wheels a little bit clumsy. The objective with this project is to find a new, better and cheaper solution that does not use expensive force sensors in the pushrims. The new power assisted wheelchair will instead only rely on its velocity, which is measured with rotational encoders, as feedback signal and thereby the project name “Torque Sensor Free Power Assisted Wheelchair”. </p><p> </p><p>The project consisted of two main parts; an extensive construction part, where an ordinary joystick controlled motorized wheelchair has been rebuild to the new power assisted wheelchair without torque sensors and a development part, where different torque sensor free controllers has been designed, simulated, programmed and tested.</p><p>The project resulted in a torque sensor free power assisted wheelchair, where the final implemented design is a proportional derivative controller, which gives a very good assisting system that is robust and insensitive to measurement noise. The proportional derivative control design gives two adjustable parameters, which can be tuned to fit a certain user; one parameter is used to adjust the amplification of the user’s force and the other one is used to change the lasting time of the propulsion influence.</p><p>Since the new assisting control system only relies on the velocity, the torque sensor free power assisted wheelchair will besides giving the user assisting power also give an assistant, which pushes the wheelchair, additional power. This is a big advantage compared to the pushrim activated one, where this benefit for the assistant is not possible.</p>

A power assisted wheelchair combines human power, which is delivered by the arms through the pushrims, with electrical motors, which are powered by a battery. Today’s electric power assisted wheelchairs use force sensors to measure the torque exerted on the pushrims by the user. The force sensors in the pushrims are rather expensive and this approach also makes the wheels a little bit clumsy. The objective with this project is to find a new, better and cheaper solution that does not use expensive force sensors in the pushrims. The new power assisted wheelchair will instead only rely on its velocity, which is measured with rotational encoders, as feedback signal and thereby the project name “Torque Sensor Free Power Assisted Wheelchair”. The project consisted of two main parts; an extensive construction part, where an ordinary joystick controlled motorized wheelchair has been rebuild to the new power assisted wheelchair without torque sensors and a development part, where different torque sensor free controllers has been designed, simulated, programmed and tested. The project resulted in a torque sensor free power assisted wheelchair, where the final implemented design is a proportional derivative controller, which gives a very good assisting system that is robust and insensitive to measurement noise. The proportional derivative control design gives two adjustable parameters, which can be tuned to fit a certain user; one parameter is used to adjust the amplification of the user’s force and the other one is used to change the lasting time of the propulsion influence. Since the new assisting control system only relies on the velocity, the torque sensor free power assisted wheelchair will besides giving the user assisting power also give an assistant, which pushes the wheelchair, additional power. This is a big advantage compared to the pushrim activated one, where this benefit for the assistant is not possible.

Background: A squat is a common exercise that is used in many areas of strength training and for different purposes and the literature is inconclusive when it comes to whether the weight lifting belt (WB) affects performance and/or is injury-preventing. The use of breathing techniques is common during heavy lifting and therefore the practice of the breathing teqnice; valsalva maneuver (VM) may be of interest to study and if this along with the WB can provide some advantages in power output and velocity. Aim: The specific aim of the study was to evaluate whether the velocity in the eccentric and the concentric phase of the squat, and the peak velocity in the concentric phases are affected in power output through the use of the VM when the subjects use or did not use a WB. Method: Fifteen subjects (10 men and 5 women) volunteered freely to participate and did a total of 12 squats divided in four different sets with three repetitions each on 75% of their self-reported one repetition maximum (1RM). The first two sets were either with or without WB and the third and fourth sets were either with or without the practice of the VM. The three conditions (with WB, with WB + VM and VA only) were compared to each other and to the control group (without any instructions and no WB) in terms of power output and velocity in the eccentric, concentric and peak velocity in the concentric phase of the squat. Result: There was no significant difference in power output when comparing the four different test conditions. The velocity in the eccentric, concentric and peak velocity in the concentric phase did not have a significant difference between the different test conditions. Conclusions: This study shows a different output compared to previous literature. The WB and the practice of VM did not affect the power output and velocity in a squat, alone or together. / Bakgrund: Det finns många olikheter i litteraturen när det gäller huruvida tyngdlyftarbältet påverkar prestationen och/eller om det minskar skaderisken. En knäböj är en vanlig övning som används inom många områden av styrketräning och för olika ändamål. Användning av andningstekniker är vanligt vid tunga lyft och därför kan utförandet av andningstekniken; valsalvamanövern vara av intresse att studera och om det tillsammans med lyftbältet kan ge effekt på effektutveckling och hastighet i lyft. Syfte: Syftet med studien var att utvärdera hastigheten i en knäböjs olika faser (excentriska, koncentriska och topphastigheten i den koncentriska fasen) och hur effektutvecklingen påverkas av lyftarbälte och valsalvamanövern. Metod: Femton personer (10 män och 5 kvinnor) deltog frivilligt och utförde totalt 12 knäböj i fyra olika sets med tre repetitioner på 75 % av testpersonernas självrapporterade 1RM. De första två seten var utförda antingen med eller utan tyngdlyftarbälte och de tredje och fjärde seten var utförda antingen med eller utan utövande av valsalvamanövern. Dessa tre förhållanden ( med lyftarbälte, med lyftarbälte + VA och VA endast) jämfördes med varandra och med kontrollgruppen ( ingen VM och inget lyftarbälte) med avseende på effektutveckling och hastigheten i den excentriska, koncentriska och topphastighet i knäböjens koncentriska fas. Resultat: Effektutvecklingen gav ingen signifikant skillnad i någon av de fyra olika förutsättningarna (med lyftarbälte, utan lyftarbälte, med bälte och valsalvamanövern och utan bälte och valsalvamanövern). Hastigheten i den excentriska, koncentriska och topphastigheten i den koncentriska fasen visade ingen signifikant skillnad mellan de fyra olika seten. Konklusion: Studien visade ingen skillnad vilket kan jämföras med tidigare litteratur där en skillnad fanns. Lyftarbältet och utförandet av valsalva manövern påverkade inte effektutvecklingen och/eller hastigheten när en knäböj utfördes.

The goal of this thesis is to describe and build a system for neural machine translation. System is built with recurrent neural networks - encoder-decoder architecture in particular. The result is a nmt library used to conduct experiments with different model parameters. Results of the experiments are compared with system built with the statistical tool Moses.

This master's thesis deals with lossless video compression. This thesis includes basic concepts and techniques used in image representation. The reader can find an explanation of basic difference between lossless video compression and lossy video compression and lossless video compression limitations. There is also possible find a description of the basic blocks forming the video codec (every block is described in detail). In every block there are introduced its possible variants. Implemented lossless videocodec was compared with common lossless videocodecs.

<p>Disertacija istražuje primenu fleksibilne elektronike za<br />kapacitivne senzore ugla i ugaone brzine tipa apsolutnog i<br />inkrementalnog enkodera cilindrične strukture. Razmatraju se dve<br />strukture, apsolutnog i inkrementalnog enkodera. Izvršena je analiza<br />uticaja mehaničkih nesavršenosti na funkciju kapacitivnosti.<br />Razvijena su dva prototipa kapacitivnih senzora za statičko i<br />dinamičko ispitivanje karakteristika senzora. Razvijena je<br />elektronika za obradu senzora inkrementalnog tipa sa<br />autokalibracijom senzora.</p> / <p>In this thesis a research on application of flexible electronics for capacitive<br />angular position and speed sensors, referred to as absolute and incremental<br />encoders, is done. It considers two structures of absolute and incremental<br />encoder type. An analysis of mechanical inaccuracies influence on a<br />capacitance function is conducted. Two prototypes are developed and used<br />for static and dynamic measurements of capacitive sensor&#39;s characteristics.<br />An electronics front-end for a capacitive two channel incremental encoder with<br />auto-calibration is developed.</p>

Patent monitoring is about tracking the upcoming inventions in a particular field, predicting future trends, and specific intellectual property rights of interest. It is the process of finding relevant patents on a particular topic based on a specific query. With patent monitoring, one can keep them updated on the new technology in the market. Also, they can find potential licensing opportunities for their inventions. The outputs of patent monitoring are essential for companies, academics, and inventors looking forward to using the latest patents that can enhance further innovation. Nevertheless, there is no widely accepted best approach to patent monitoring. Usually, most patent monitoring systems are based on complex search and find, often leading to insignificant hit rates and highly human intervention. As the number of patents published each year increases massively and with patents being critical to accelerating innovation, the current approach to patent monitoring has two main drawbacks. Firstly, human-driven patent monitoring is time consuming and expensive process. In addition, there is a risk of overlooking interesting documents due to inadequate searching tools and processes, which could cost companies fortunes while at the same time hindering further innovation and creativity. This thesis presents a smart patent monitoring assistant tool that applies natural language processing. The use of several natural language processing methods is investigated to find, classify and rank relevant documents. The tool was trained on a dataset that contains the title, abstract, and claims of patent documents. Given a dataset of patent documents, the aim of this thesis is to create a tool that can classify patents into two classes relevant and not relevant. Furthermore, the tool can rank documents based on relevancy. The evaluation result of the tool gave satisfying results when it came to receiving the expected patents. In addition, there is a significant improvement in terms of performance for memory usage and the time it took to train the model and get results. / Patentövervakning handlar om att övervaka kommande uppfinningar, förutsäga framtida trender, eller specifika immateriella rättigheter och används för att hitta relevanta patent inom ett visst område. Med patentövervakning är det möjligt att hålla patent uppdaterade enligt den senaste tekniken på marknaden samt att hitta potentiella möjligheter att licensiera innehavda patent till tredje part. Målgruppen för patentövervakning är företag, akademiker, och uppfinnare som vill hitta de senaste patenten för att uppnå maximal innovation. Dock finns det ingen generell metod för att bedriva patentövervakning. Vanligtvis används komplexa sökmetoder som resulterar i undermåliga resultat och kräver manuellt ingripande. I och med att andelen patent ökar varje år har nuvarande metod två huvudsakliga nackdelar. Till att börja med är mänsklig patentövervakning en tidskrävande och dyr process. Vidare är det en betydande risk att missa viktiga eller på andra sätt intressanta dokument till följd av en bristande sökprocess. Detta kan möjligtvis resultera i att företag missar stora möjligheter samt utebliven innovation och kreativitet. Detta arbete presenterar ett smart verktyg för patentövervakning baserat på natural language processing. Vi analyserar användningen av ett flertal processer för att hitta, klassificera, och rangordna relevant dokument. Verktyget tränades på ett dataset som innehåller patentets titel, abstrakt, och vad patentet gör anspråk på. Givet ett godtyckligt dataset är målet med detta arbete att utveckla ett verktyg med förmågan att klassificera relevanta och icke-relevanta patent samt rangordna dessa utifrån relevans. Resultatet visar att verktyget gav tillfredsställande gällande att hitta önskvärda patent. Vidare uppnåddes en signifikant förbättring när det gäller prestanda för minnesanvändning och tiden som krävs för att träna modeller och erhålla resultat.

For a long time stereo-cameras have been deployed in visual Simultaneous Localization And Mapping (SLAM) systems to gain 3D information. Even though stereo-cameras show good performance, the main disadvantage is the complex and expensive hardware setup it requires, which limits the use of the system. A simpler and cheaper alternative are monocular cameras, however monocular images lack the important depth information. Recent works have shown that having access to depth maps in monocular SLAM system is beneficial since they can be used to improve the 3D reconstruction. This work proposes a deep neural network that predicts dense high-resolution depth maps from monocular RGB images by casting the problem as a supervised regression task. The network architecture follows an encoder-decoder structure in which multi-scale information is captured and skip-connections are used to recover details. The network is trained and evaluated on the KITTI dataset achieving results comparable to state-of-the-art methods. With further development, this network shows good potential to be incorporated in a monocular SLAM system to improve the 3D reconstruction.