A convolutional neural network (CNN) is a deep learning framework that is widely used in computer vision. A CNN extracts important features of input images by perform- ing convolution and reduces the parameters in the network by applying pooling operation. CNNs are usually implemented with programming languages and run on central process- ing units (CPUs) and graphics processing units (GPUs). However in recent years, research has been conducted to implement CNNs on field-programmable gate array (FPGA). The objective of this thesis is to implement a CNN on an FPGA with few hardware resources and low power consumption. The CNN we implement is for digits recognition. The input of this CNN is an image of a single digit. The CNN makes inference on what number it is on that image. The performance and power consumption of the FPGA is compared with that of a CPU and a GPU. The results show that our FPGA implementation has better performance than the CPU and the GPU, with respect to runtime, power consumption, and power efficiency.

How can Linux reduce the time required for development in embedded systems, and what makes Linux appealing to embedded developers despite the loss in overall control? Through qualitative interviews with industry professionals and a systematic literature review, challenges and benefits of using Linux in embedded systems development were identified and discussed. Three hypotheses were formulated based on recurring topic agreement among the interview subjects: Reduced development time through the use of open-source solutions, struggles with real-time and security requirements, and challenges within troubleshooting and dependency management. The empirical data observed primarily aligned with the professional perception indicating the potential for development time reduction leveraging resources properly. However also highlighting additional challenges that are not present in traditional embedded system development. Several trade-offs were observed from the findings, including increased overhead and licensing concerns. Further research is required to fully understand the advantages, challenges and limits associated with Linux in an embedded system environment. This study provides valuable insights for future exploration within the field.

The introduction of smart grid technology has changed the way traditional power grid network function. It made the grid structure more dynamic by enhancing electrical usage management capability. Also, it has increased the scope to enhance communication infrastructure in a smart grid structure. The current smart grid solution is based on IEC 61850 architecture where the exchange of information between the electrical utilities is over the fast Ethernet LAN connection. This communication mechanism is fast, efficient but lacks scalability, flexibility and less susceptible to failure. Also, earlier technical paper from IEC 61850 standard was for communication within a substation. Wide Area Monitoring Protection and Control implementation which utilizes coherent real time synchrophasor information would play a vital role in realizing the utility physical status. IEC 61850-90-5, a new technical report from International Electrotechnical Commission provides the mechanism to transmit and receive the synchrophasor information using the advance IP protocol over a wireless communication infrastructure for WAMPAC application. IEC 61850-90-5 provide a way to exchange routable synchrophasor information over public IP network such as LTE, WiMax, WLAN, etc. Out of all the available wireless solution, LTE provides high flexibility, distance coverage, data rate with low latency and hence can play an important role in replacing the existing communication structure in a smart grid. The thesis work evaluates the performance and applicability of LTE for smart grid communication. An IEC 61850-90-5 communication model utilizing UDP/IP protocol to transmit and receive data over the LTE network was developed from the open source project. The modified model was used to benchmark the performance of LTE. Different communication metrics such as reliability, availability, latency, and throughput was evaluated to benchmark the performance of LTE for time critical smart grid application. The metrics were measured for different packet sizes and transmission rates combination. The result shares some interesting findings on the readiness of LTE for smart grid solution. It is concluded that cellular network can play an important role in realizing communication infrastructure in a smart grid application.

Controller Area Network (CAN) is a communication bus designed in the 90s to make simple androbust in-vehicle networks. However, as vehicles are becoming more complex, higher performanceCAN protocols were introduced to manage the expanding volumes of real-time data and more strictsafety requirements. This resulted in the development of next-generation CAN protocols, i.e. CANwith Flexible Data rate (CAN FD) and CAN Extra Long (CAN XL). Response-Time Analysis(RTA) of CAN was developed as a tool to verify if all messages on the CAN bus meet their timingrequirements, i.e. meet their deadlines. Existing RTA is revisited and its applicability on CAN XLmessages is analyzed in the thesis. RTA with priority-based queueing policy, but also RTA whichconsiders different buffer limitations is also revisited. This thesis developed an analytical modelfor Worst-Case Transmission Time (WCTT) calculations for CAN XL which is a prerequisite forits RTA. Finally, MPS-CAN Analyzer and industrial tool suit Rubus-ICE are extended to supportRTA of both CAN FD and CAN XL. The extended tools were used to implement two use casesconsisting of the number of CAN, CAN FD and CAN XL messages. Several experiments wereconducted to compare the performance of different CAN generations in terms of response times ofthe messages. The results indicate that larger data payloads should be sent using next-generationCAN protocol with their bit-rate switching option enabled. Nodes implementing abortable and nonabortablebuffers were also analyzed in the two experiments. Small differences in the response timesof the messages were noticed when buffer limitations are considered.

Energy systems need to evolve to meet the requirements of the modern world and the future. Hence, substantial effort is needed at an academic and industrial level to develop valuable diagnostic techniques. One current limitation of developing these diagnostic techniques is a lack of data on significant changes in the power signal (voltage and current). Producing sufficient data would require the deployment of data acquisition units that can record events of interest. These data acquisition units could be embedded systems deployed in the field. One way to detect changes in the signal is via a waveform prediction algorithm. Hence, waveform prediction will play a significant role in developing the trigger mechanism for monitoring algorithms as predicting a waveform a few cycles ahead allow the triggering of desired events. This thesis aims to find suitable machine-learning models for online and real-time time series prediction that predict the nominal state of a power signal waveform and could be used to detect significant events to activate a trigger mechanism in data acquisition devices. In this thesis, three different machine-learning models: Accurate Online Support Vector Regression (AOSVR), online Autoregressive Integrated Moving Average (ARIMA), and online Gated Recurrent Units (GRU) are implemented on a Raspberry Pi 3B+ and compared to each other in the aspects of computation time and error. Rigorous testing results demonstrate that the AOSVR had the lowest average operation in all the tested sampling rates, while the ARIMA had the overall lowest prediction time and lowest error when tested on data with sampling rates up to 2.5 kHz. Furthermore, all the models were able to produce a trigger at significant events. The results of this thesis could be used by the industry to further the development of diagnostic techniques, as well as increase the general knowledge of machine-learning applications on embedded systems.

No description available.

The development of neural networks has become one of the most interesting topics in the scientific community. Systems that are based on the brain behavior can find applications in a wide variety of fields, from simulating the brain to better understand it (applications in neuroscience), to control theory and super computing. Brain-like systems could possible be a new kind of computer architecture that will lead us away from the classic von Neumann architecture. That can help us bypass the problems that we now face, with the Moore’s low slowing down and complex problems becoming all the more common. With brain-like computing, we might be in the road to computer systems that are no longer programmed but taught. To-date, the most common platform for simulating such systems are the GPGPUs and super computers. But they lack on scalability, and real time simulations are far from trivial. Because of that there is an interest in custom hardware implementation of such system (in ASIC or FPGAs). In this work, we focus on the ASIC design of such a system. Specifically, with the characterization and design space exploration of the eBrain architecture, a hardware architecture for the BCPNN model. During the design process of an ASIC, in order to be able to characterize it, the simulation of the synthesized physical design of the RTL model is required. Those kinds of simulations require an extensive amount of time. In this thesis, to tackle with this problem a systemC model of the architecture is developed. This model is able to be modified and fits different configurations of a general hardware architecture. The systemC model can be used to reduce the amount of time the simulation requires and, by using back annotated data from synthesized parts of the hardware architecture, to provide us with accurate characterization of the design. In this work, we go through the basics of the BCPNN and the eBrain architecture. Then we develop a model that can emulate the behavior of the eBrain architecture in a probabilistic manner. A specific configuration is chosen to be explored. Furthermore, floating-point units are synthesized in the physical level in order to be able to back annotate their power measurements to the model. Moreover, the BCPNN equations are explored and implemented in an RTL level with the use of the floating-point units. Finally, an example configuration is simulated and its results are presented.

Neuromorphic computing is a recent and growing field of research. Its conceptual attractiveness is due to the potential it has in deep learning applications such as sensor networks, low-power computer vision, robotics and other fields. Inspired by the functioning of brain, different neural network models have been devised, each with their own special focus on certain applications. Using such computing models are already helping us in different cases such as image, character and voice recognition, data analysis, stock market prediction, etc. Among the multitude of artificial neural models available, spiking neurons are more deeply inspired by biological neural networks. Leaky, Integrate and Fire (LIF) neuron model is one such model that can reproduce a good number of functions, be simple and also extensible in structure. Current deep learning applications are tied to servers and datacenters for their power and resource hungry existence. This work aims at building a low power neuron core taking advantage of LIF neuron, that could possible result in independent battery powered devices. A hardware design of LIF neuron based scalable neural core is explored, constructed and analysis for power consumption is made.

The need for electromagnetic compatibility is growing steadily as the usage of electronics in our daily lives is increasing more than ever. A common issue encountered in electromagnetic compatibility testing is analogue sensors that fail when exposed to electromagnetic fields. Testing how well electronics do when exposed to electromagnetic fields is called susceptibility testing, and standards for how to do these tests have been developed to ensure that the results of the tests can be reproduced. In this thesis work, analogue temperature sensors have been shielded using a few common techniques. The susceptibility of the sensors has been analysed by looking at their output voltage when the sensors were exposed to electro- magnetic fields of different field strengths. The output of the sensors was read by an Arduino that was shielded and tested to make sure it would not be affected by the electromagnetic fields used in the sensor tests. The result of the first set of sensor tests shows that shielding the cables running to the analogue temperature sensors and filtering away disturbances using ferrites gives a considerable decrease in susceptibility against electro- magnetic fields, while twisted cables and RC-filters did not. The results also showed that the introduction of a ground plane increased the susceptibility of the sensors, which most likely was due to it not providing the current with a path of less impedance and only served to increase the length of the unintentional antenna, which made it couple to the electromagnetic field more easily. However, during a second round of testing, the results of all the tests were hard to reproduce exactly, which calls into question how trustable the results of standardised susceptibility tests are. Therefore, when designing for the electromagnetic susceptibility of a product, a rather wide margin should be used in order to make sure that the product can reliably pass susceptibility tests. / Användningen av elektronik ökar i samhället och därför även nödvändigheten för testning av elektromagnetisk kompatibilitet. Ett vanligt problem inom elektromagnetisk kompatibilitet är att analoga sensorer lätt blir utstörda av elektromagnetiska fält. Hur man ska testa en elektronisk produkts känslighet mot elektromagnetiska fält styrs av standarder som ser till att resultaten av testerna går att återskapa. I detta examensarbete har analoga temperatursensorer skärmats med ett par vanliga metoder. Sensorernas känslighet har analyserats genom att undersöka hur deras utspänning påverkas när sensorn blir utsatt för elektromagnetiska fält med olika fältstyrkor. Sensorernas utspänning lästes av en Arduino som skärmades och testades för att se till all att den inte påverkades av de elektromagnetiska fälten som användes under testandet av sensorerna. Resultaten från de första sensortesterna visar att använda skärmade kablar till de analoga temeperatursensorerna och att filtrera bort störningar med ferriter sänkte sensorernas känslighet mot elektromagnetiska fält betydligt medan tvinnade kablar och RC filter inte gjorde det. Testerna visade också att jord- plan i detta fall ökade sensorernas känslighet då de inte erbjöd en bättre väg för strömmen att gå utan endast skapade en längre oavsiktlig antenn, vilket gjorde att den lättare kunde koppla till det elektromagnetiska fältet. Däremot visade det sig i en andra testomgång, att resultaten inte gick att återskapa ex- akt. Detta ifrågasätter hur tillförlitliga dessa standardiserade tester är och visar att man bör ha en ganska bred marginal när man designar för att minska en produkts känslighet mot elektromagnetiska fält, så att den på ett tillförlitligt sätt kommer kunna klara av känslighetstester.

Wireless power and data transfer systems are experiencing an ever-growing consumer and industrial adoption. Its use in common devices has made the technology more accessible to people, but seldomly does it replace a physical connection inside a product. When two parts of an assembly are to be electrically connected, the solution has traditionally been connectors or wires. However, typical connection methods using physical connectors between two devices can be fragile and sensitive to dust and debris. In this degree project, a wireless power and data connection between two parts of an industrial nutrunner are evaluated. A very compact nutrunner encasement calls for a minimal wireless interface with high efficiency. Additional complications are met when the nutrunner body is made of metal, which introduces losses. Electromagnetic simulations of a flexible PCB transformer with ferrite film backing are performed in the simulation software Finite Element Method Magnetics (FEMM), and electrical performance is evaluated in the circuit simulator LT-Spice. From the best performing solution, a physical model is constructed and evaluated. The final implementation uses a flyback converter for power transfer, and amplitude modulated data for bi-directional data transfer. Results indicate the potential for excellent performance with 1 W power transfer with more than 50 % efficiency whilst simultaneously transferring data at a rate greater than 1 Mbit/s. / Trådlös effekt och dataöverförings system ser allt fler tillämpningar i konsument och industriella sektorn. Dess implementation i vardagliga enheter har gjort teknologin mer tillgänglig för människor, men den ersätter sällan fysiska kontakter inuti produkter. När två delar av en produkt skall sammankopplas elektriskt har den traditionella lösningen varit kontakter eller sladdar. Däremot så kan typiska anslutningsmetoder med fysiska kontakter vara ömtåliga samt känsliga för damm och smuts. I detta examensarbete undersöks ett trådlöst effekt och data gränssnitt mellan två delar av en industriell mutterdragare. En väldigt kompakt mutterdragrinkapsling medför ett väldigt kompakt trådlöst gränssnitt med hög verkningsgrad. Ytterligare komplikationer uppstår när mutterdragaren är gjord av metall, vilket medför förluster. Elektromagnetiska simuleringar av en flexibel kretskorts transformator utförs i simuleringsprogrammet Finite Element Method Magnetics (FEMM) och elektrisk prestanda undersöks i kretssimulatorn LT-Spice. En prototyp konstrueras från den lösning som presterat bäst och denna utvärderas. Den slutgiltiga implementationen använder en flyback omvandlare för effektöverföring och amplitud modulerade data för dubbelriktad datakommunikation. Resultat indikerar god prestanda med 1 W effekt överfört vid mer än 50 % verkningsgrad samtidigt som data överförs med en hastighet mer än 1 Mbit/s.