No description available.

As the world becomes more globalised, user equipment such as smartphones and Internet of Things devices require increasingly more data, which increases the demand for wireless data traffic. Hence, the acceleration of next-generational networks (5G and beyond) focuses mainly on increasing the bitrate and decreasing the latency. A crucial technology for 5G and beyond is the massive MIMO. In a massive MIMO system, a detector processes the received signals from multiple antennas to decode the transmitted data and extract useful information. This has been implemented in many ways, and one of the most used algorithms is the Zero Forcing (ZF) algorithm. This thesis presents a novel parallel design to accelerate the ZF algorithm using the Cholesky decomposition. This is implemented on a GPU, written in the CUDA programming language, and compared to the existing state-of-the-art implementations regarding latency and throughput. The implementation is also validated from a MATLAB implementation. This research demonstrates promising performance using GPUs for massive MIMO detection algorithms. Our approach achieves a significant speedup factor of 350 in comparison to a serial version of the implementation. The throughput achieved is 160 times greater than a comparable GPU-based approach. Despite this, our approach reaches a 2.4 times lower throughput than a solution that employed application-specific hardware. Given the promising results, we advocate for continued research in this area to further optimise detection algorithms and enhance their performance on GPUs, to potentially achieve even higher throughput and lower latency. / <p>Our examiner Mahdi wants to wait six months before the thesis is published. </p>

This report presents the design of a conceptual prototype aimed at identifying keys in specific positions inside a cabinet utilizing Radio-frequency identification (RFID)technology. The prototype integrates RFID readers, managed by a microcontroller unit(MCU), establishing a backend peripheral system. The cabinet is made of steel, and given RFID’s sensitivity to nearby metal, experimentation was conducted to evaluate the impact of metal proximity on the reading range. Experimental results reveal a reduction in the reading range of 15 mm (43%) by one metal sheet and 26 mm (74%) by two metal sheets present, highlighting the relation between RFID technology and metallic environments. Additionally, the finished prototype is also presented in the Results and Discussion section, giving a more detailed insight into its practical implementation. This project demonstrates the viability of item-level identification through the utilization of low-frequency readers. Particularly relevant for positional identification, the short reading range of a low-frequency reader offers precision by limiting the area in which a detected transponder may be located. / Denna rapport presenterar designen av en konceptuell prototyp avsedd att identifiera nycklar på specifika positioner i ett skåp med användning av Radio-frequency identification (RFID) teknologi. Prototypen integrerar RFID läsare, hanterade av en microcontroller unit (MCU), och etablerar ett perifert backendsystem. Skåpet är gjort av stål, och givet RFIDs känslighet mot nära metal har experiment utförts för att utvärdera hur metallen påverkar en läsares omfång. Experimentella resultat avslöjar en reduktion i läsarens omfång på 15 mm (43%) med en metallplatta och 26 mm (74%) med tvåmetallplattor närvarande. Dessutom presenteras den färdiga prototypen i Resultat och Diskussionsdelen, för att ge en mer detaljerad insyn på dess praktiska implementation. Detta projektet demonstrerar genomförbarheten av identifikation på artikelnivå genomanvändandet av lågfrekvens läsare. Särskilt relevant för positionell identifikation är det korta omfånget av en lågfrekvens läsare, då detta erbjuder precision genom att begränsaområdet där en transponder kan lokaliseras.

The Internet of Things is deploying large numbers of bare-metal systems that have no protection against memory corruption and control-flow hijacking attacks. These attacks have enabled unauthorized entry to hotel rooms, malicious control of unmanned aerial vehicles, and invasions of privacy. Using static and dynamic analysis these systems can utilize state-of-the-art testing techniques to identify and<br>prevent memory-corruption errors and employ defenses against memory corruption and control-flow hijacking attacks in bare-metal systems that match or exceed those currently employed on desktop systems. This is shown using three case studies.<br><br>(1) EPOXY which, automatically applies data execution prevention, diversity, stack defenses, and separating privileged code from unprivileged code using a novel<br>technique called privileged overlaying. These protections prevent code injection attacks, and reduce the number of privileged instruction to 0.06% verses an unprotected<br>application.<br><br>(2) Automatic Compartments for Embedded Systems (ACES), which automatically creates compartments that enforce data integrity and code isolation within bare-metal applications. ACES enables exploring policies to best meet security and performance requirements for individual applications. Results show ACES' can form 10s of compartments within a single thread and has a 15% runtime overhead on average.<br><br><div>(3) HALucinator breaks the requirement for specialized hardware to perform bare-metal system testing. This enables state-of-the-art testing techniques –e.g., coverage based fuzzing – to scale with the availability of commodity computers, leading to the discovery of exploitable vulnerabilities in bare-metal systems. <br></div><div><br></div><div>Combined, these case studies advance the security of embedded system several decades and provide essential protections for today’s connected devices.</div>

A demonstration robot for conferences and fairs has been built from scratch. The demonstration robot is meant to create lasting impressions at the company booth at conferences or fairs. Thus, the robot needs traits that attract people to the booth and makes sure they remember that company. In this project, traits such as being able to move, do facial expressions and play audio have been developed. The robot has also been designed to draw as much attention as possible to the booth. This was achieved by building a robot that consists of a rolling sphere with a head that always remains on top. All movements are carried out from inside the sphere by four different motors. One motor moves the robot back and forth, two motors spin a flywheel to turn the robot and the last motor rotates the head. These motors are mounted at different places on an internal structure. The internal structure is connected to the sphere at two points, one on each side of the robot. At the top of the internal structure, magnets are placed. Thus, it can attach the head at the outside of the sphere by mounting magnets in the head. All movements of the robots are controlled by a hand controller, which has been made in this project. The head has a built-in display simulating two eyes. The display is driven by a Raspberry Pi. An internal speaker is built-in inside the head, connected to the Raspberry Pi. Each simulated eye consists of 64 squares that can be programmed to be in different colours, thus making it possible to express a wide range of facial expressions.Two PCBs were designed and manufactured to control the robot. One was placed inside the robot, and the other inside the hand controller. The PCBs can communicate over Bluetooth, which makes it possible to control the robot from the outside.All parts of the robot have been designed in a CAD program and subsequently 3D printed. 3D design in CAD was learned from a novice level, since there was no previous knowledge of this in the project group.In addition, a registration form has been developed that allows visitors to register at the booth. Making it easier for the company to connect with visitors after the conference or fair. The registration form is connected with the Raspberry Pi in the head of the robot via Wi-Fi. Thus, when new registrations occur, the robot can print the names of them while audio is played.

Software systems in the automotive domain are generally safety critical and subject to strict timing requirements. Systems of this character are often constructed utilizing periodically executed tasks, that have a hard deadline. In addition, these systems may have additional deadlines that can be specified on cause-effect chains, or simply task chains. They are defined by existing tasks in the system, hence the chains are not stand alone additions to the system. Each chain provide an end-to-end timing constraint targeting the propagation of data through the chain of tasks. These constraints specify the additional timing requirements that need to be fulfilled, when searching for a valid schedule. In this thesis, an offline non-preemptive scheduling method is presented, designed for single core systems. The scheduling problem is defined and formulated utilizing Constraint Programming. In addition, to ensure that end-to-end timing requirements are met, job-level dependencies are considered during the schedule generation. Utilizing this approach can guarantee that individual task periods along with end-to-end timing requirements are always met, if a schedule exists. The results show a good increase in schedulability ratio when utilizing job-level dependencies compared to the case where job-level dependencies are not specified. When the system utilization increases this improvement is even greater. Depending on the system size and complexity the improvement can vary, but in many cases it is more than double. The scheduling generation is also performed within a reasonable time frame. This would be a good benefit during the development process of a system, since it allows fast verification when changes are made to the system. Further, the thesis provide an overview of the entire process, starting from a system model and ending at a fully functional schedule executing on a hardware platform.

The car is the most common vehicle in the world. Millions of cars are produced annually. In order for each car to find its buyer, car companies are forced to constantly improve the design of the car. Modern models are emerging, new car systems are being developed and implemented. All this is accompanied by a huge flow of information, in which it is easy to get lost. This master’s work is devoted to the trace analysis and connection of two different files. The paper proposes a developed algorithm of trace analysis for some functions of the vehicle in the C++ programming language. The files that we use to trace analysis relate to the model and the final result of its simulation.EATOP is a tool with which a model based on the EAST-ADL language was developed. Adapt is an event simulator with which our model of automotive functionality was simulated. The purpose of the study is to identify possible ways to meet timing requirements. The work is carried out in collaboration with Volvo Group Truck Technology. This company provided the LogFile, which presents results of the simulation, and model. We get an analysis of performance, one of the ways to trace data and timing. The results of our implementation are presented and discussed.

The increasing complexity of embedded systems development is becoming difficult to handle with development environments based on disjoint engineering tools. Support for interactions between various engineering tools, especially through automated means, has therefore received an increased amount of attention during the last few years. The subsequent increase in the amount of tool integration is leading to an increased impact of tool integration on non-functional properties of development efforts, development environments and end products. At the same time there is a lack of methods and tools for analysing the relationship between these properties and tool integration. To establish a foundation for analysing this generic relationship, the specific relationship between tool integration and the safety of end products is analysed in this thesis. A survey was conducted to analyze the State of the Art of tool integration as related to safety. This survey specifically identified the lack of an efficient handling of tool integration by modern safety standards as an important concern. In relation to this survey, three theories were identified as of specific importance. These are the school of thought known as Systems Thinking, the Systems-Theoretic Accident Model and Processes (STAMP) causality model and the System-Theoretic Process Analysis (STPA) hazard analysis technique. Building on these theories, this thesis provides original contributions intended to (1) describe concepts and models related to tool integration and safety (the first and second contribution), (2) link tool integration to safety in a way that reduces complexity during analysis (the third contribution) and (3) propose how to interpret and make use of the implications of the presented theories and the first three contributions (the fourth and fifth contribution). • The first contribution is a new conceptual model of a development effort that emphasizes tool integration. • The second contribution is a new reference model for tool integration in highly heterogeneous environments. • The third contribution consists of nine safety-related tool chain properties, i.e. properties of tool chains that could mitigate at least part of the risks introduced by tool integration. • The fourth contribution is a proposition on how to identify safety implications due to a high level of automation of tool integration. • The fifth contribution is a proposition for a new software tool qualification process.

I det här arbetet presenteras ett system som utvecklats i syfte att kunna räkna personer. Systemet är tänkt att användas i mötesrum för upp till tio personer och använder sig av infraröd teknik i form av thermopile arrayer. I arbetet har tre olika sensorer använts för utvärdering. Sensorerna som använts är Panasonic Grid-EYE med pixelupplösningen 8×8 och detekteringsvinkeln 60°×60° samt två stycken Heimann 32x32d, båda med upplösningen 32×32 pixlar, men med detekteringsvinkeln 90°×90° respektive 105°×105°. Systemet är programmerat med hjälp av utvecklingskortet STM32L476RG och är skrivet i språket C. I systemet används två metoder för att beräkna antalet personer. Den ena metoden jämför temperaturförändringen i ett rum i förhållande till då rummet är tomt och den andra metoden använder sig av bildbehandlingsmetoder som interpolering, filtrering och beräkning av area. Sensorerna utvärderas även individuellt utifrån egenskaper som noggrannhet, strömförbrukning och implementationskostnad. Script har även skapats i MATLAB som, i kombination med mikrokontrollern, används för att grafiskt presentera temperaturvärdena från sensorerna. Den sensor som visade sig vara bäst lämpad för att räkna personer är Heimann 32×32d med detekteringsvinkeln 105°×105°. Detta tack vare den större detekteringsvinkeln som resulterar i en större detekterbar yta samt upplösningen på totalt 1024 pixlar som sammantaget ger en högre noggrannhet för personräkning. Denna sensor kräver dock mer komplexa och tidskrävande beräkningar för behandling av data än Grid-EYE. Dessa skillnader är ändå marginella, där noggrannheten och den större detekterbara arean väger upp nackdelarna. De experimentella resultaten visar att Heimann 32×32d med 105°×105° ger en noggrannhet på c:a 98.3 % vid mätning på höjden 2.45 m. Detta motsvarar en yta på c:a 39.1 m2 och systemet kan räkna upp till minst 4 personer. För Grid-EYE och samma höjd har ej noggrannheten fastställts, men har endast en detekterbar yta på c:a 7.7 m2 där maximalt 4 personer bedöms kunna räknas. / In this work, a system is being developed with the purpose of counting people. The system is intended for use in meeting rooms for up to ten persons and utilizes infrared technique using thermopile arrays. For this work, three different sensor have been used for evaluation. A Panasonic Grid-EYE with a resolution of 8×8 and a Field of View (FoV) of 60°×60°, and two Heimann 32×32d sensors, both having a resolution of 32×32, but with the FoV 90°×90° and 105°×105° respectively. The system has been programmed using the microcontroller STM32L476RG, and with the programming language C. In this system, two methods for people counting has been implemented. The first method compares the total change in temperature of a room in relation to when the room is empty and the other method uses image processing methods, such as interpolation, filtering and area calculations. The sensors are also being evaluated individually, based on characteristics such as accuracy, current consumption and implementation cost. To graphically display the temperature values of the sensors, scripts has been made for MATLAB that uses information sent by the microcontroller. The sensor which seem to be best suited for counting people is the Heimann 32×32d with the FoV 105°×105°. The main advantage of this sensor is its wider FoV, covering a larger area, and its higher resolution, which overall yields a higher accuracy when counting people. However, this sensor requires more complex and time-consuming calculations when processing data than the Grid-EYE. Still, these differences are marginal where the accuracy and the larger detectable area for the HTPA outweighs its disadvantages. The experimental results shows that the Heimann sensor with 105°×105° FoV can achieve an accuracy of ≈98.3 % measuring at a height of 2.45 m. At this height, the detectable area for the sensor is ≈39.1 m2 and is being able to count up to at least 4 persons. As for the Grid-EYE and with the same scenario, the accuracy has not been determined, but has a detectable area of ≈7.7 m2 and is estimated being able to count up to a maximum of 4 persons.

Att köra fastän trötthet upplevs är en erkänd trafikfara. Om en förare idag kör när denne är trött klassas det som vårdslöshet i trafiken. Dock är det nästan omöjligt att vid en olycka påvisa att olyckan skedde för att en förare var trött. Det praktiska resultatet är att det lämnas åt föraren att själv avgöra om denne är för trött för att föra fram fordonet säkert. Hela situationen om att vara en trafikfara eller ej lämnas då åt en subjektiv bedömning.   I denna rapport undersöks olika sömnstadier, och metoder för att avgöra när och hur det är lämpligast att vakna för att vakna i ett mer alert tillstånd. För att sedan använda de metoder som visar sig vara lämpade som indata och parametrar till en algoritm som ska avgöra när det är som lämpligast att vakna för att vakna mer alert. Rapporten gör också en kalkyl som kan användas för att fastställa prototypens utvecklingskostnader.   Arbetet resulterar i en prototyp som föreslår en lösning till problemet med trötthet i trafiken. De kostnader som framkommer i samband med prototyputveckling fastställs och presenteras.   Undersökningen och utvecklandet av den tekniska lösningen har delats upp i två moduler och kommer att göras i samverkan med en extern studentgrupp på skolan för elektroteknik och datavetenskap (EECS) vid Kungliga Tekniska högskolan. / Driving although experiencing fatigue is a recognized traffic hazard. Today, if a driver is driving despite being tired, it is considered to be reckless driving. However, in the event of an accident, it is almost impossible to prove that the accident occurred because the driver was tired. The practical result is that it is left to the driver to decide if one is too tired or not to drive the vehicle safely. The whole situation of being a traffic hazard or not is then subjected to a subjective assessment. This report examines different sleep stages and methods for determining when it’s most appropriate, and how it’s most appropriate to wake up, with the aim on waking up in a more alert state. Further, to use the methods that prove to be suitable as inputs and parameters of an algorithm. The report also examines different calculations that can be used to determine the development cost of the prototype. The result consists of a prototype that is a beginning on addressing the problem with fatigue in traffic. It also consist of a calculation determining the development cost. The research and development of the technical solution has been divided into two modules and will be done in collaboration with an external student group at School of Electrical Engineering and Computer Science (EECS) at KTH Royal Institute of Technology.