Introduction: Software reuse by copying and modifying components to fit new systems is common in industrial settings. However, it can lead to multiple variants that complicate testing and maintenance. Therefore, it is beneficial to detect the variants in existing codebases to document or incorporate them into a systematic reuse process. For this purpose, model-based clone detection and variability management can be used. Unfortunately, current tools have too high computational complexity to process multiple Simulink models while finding commonalities and differences between them. Therefore, we explore a novel approach called MatAdd that aims to enable large-scale industrial codebases to be processed. Objective: The primary objective is to process large-scale industrial Simulink codebases to detect the commonalities and differences between the models. Context and method: The work was conducted in collaboration with Addiva and Alstom to detect variants in Alstom's codebase of Simulink models. Alstom has specific modeling guidelines and conventions that the developers follow. Therefore, we used an exploratory case study to change the research direction depending on Alstom's considerations. Results and Conclusions: The results show that MatAdd can process large-scale industrial Simulink codebases and detect the commonalities and differences between its models. MatAdd processed Alstom's codebase that contained 157 Simulink models with 7820 blocks and 9627 lines in approximately 90 seconds and returned some type-1, type-2, and type-3 clones. However, current limitations cause some signals to be missed, and a more thorough evaluation is needed to assess its future potential. MatAdd's current state assists developers in finding clones to manually encapsulate into reusable library components or find variants to document to facilitate maintenance.

Verification and validation of embedded systems software is tedious and time consuming. Software model checking uses a tool-based approach automating this process. In order to more accurately model software it is necessary to provide hardware support that enables the execution of software as it should run on native hardware. Hardware support often requires the creation of model checking tools specific to the instruction set architecture. The creation of software model checking tools is non-trivial. We present a strategy for using an "off-the-shelf" model checking tool, Bogor, to provide support for multiple instruction set architectures. Our strategy supports key hardware features such as instruction execution, exceptional control flow, and interrupt servicing as extensions to Bogor. These extensions work within the tool framework using existing interfaces and require significantly less code than creating an entire model checking tool.

The purpose of this thesis rose from tackling a specific project led by Cybercom Group. Water Sense was proposed after municipalities expressed the need to conveniently conduct measurements on water in Swedish lakes, and remotely get access to these data. The varying and infrequent nature of measurements result in an inefficient power consumption when a conventional sensor platform is used. Internet of Things is a relatively new and ever-evolving field for wireless sensor networks, where Low Power Wide-Area Network protocols are utilized to cleverly save power. One of these LPWAN protocols is LoRaWAN, a MAC layer protocol that runs on top of the PHY layer protocol LoRa. The chosen platform hardware for sensor nodes is an Adafruit feather equipped with a LoRa radio module, and the Contiki operative system was to be imported and integrated. Contiki OS is developed for small IoT systems with low-power, while offering networking mechanisms and a range of protocols typically utilized in WSNs. The adaptive and modular nature of Contiki allows for custom pairing of protocols to target a specific topology. LoRaWAN has a number of constraints that are disadvantageous for Water Sense, most prominently being a single-hop protocol. Contiki already supports several platforms with drivers, but the CPU and radio module of the Adafruit feather are not one of them. The goal was to integrate the LoRa PHY hardware with Contiki and have an adaptive platform for Water Sense and other scenarios. Unfortunately not all initial goals were achieved, and the physical layer was not fully integrated with Contiki’s APIs. However, the drivers needed for an operational physical layer were completed and range test could be conducted. Therefore, this thesis details the work done for implementing the physical layer and a study on Contiki’s data link protocols in proposed configurations for Water Sense.

This thesis aims to build a modular embedded system for a set of pre defined constrains, including a code algorithm for temporary access codes. The system are to be physically adapted for electronic code locks and foremost a pre-existing lock. The thesis provides a set of tools that will be used for creating the system. Furthermore the choice of components and the design decisions are accounted for. The software of the system is discussed along with the choosen security solutions. Lastly it discusses the problems with product certifications that this kind of system has to face.

Unison is a code generator integrated into the LLVM compilation chain which solves the problem of instruction scheduling and register allocation in conjunction as a constraint optimization problem, producing possibly optimal code. The solution from LLVM of the very same problem acts as a baseline for the code generated by Unison, and along with an improved version of the code is an estimation of the speedup compared to the baseline given by Unison.This thesis investigates the accuracy of this estimation for the Hexagon v4 DSP by executing the code produced by LLVM and Unison on a simulation platform. The results show that the estimations is not accurate, but there is a strong correlation between the given estimation and the observed speedup. The main reason for inaccuracy is hardware stalls of which the code produced by Unison suffers more of compared to the code produced by LLVM. / Unison är en kodgenerator integrerad i LLVMs kompileringskedja som löser problemen med instruktionsschemaläggning och registerallokering tillsammans som ett optimeringsproblem, vilket möjligen resulterar i optimal kod. Lösningen av LLVM för samma problem agerar måttstock för koden som Unison genererar, och tillsammans med en förbättrad version av koden ger Unison ifrån sig en uppskattning om hur mycket snabbare dess kod är.Den här avhandligen undersöker precisionen för denna uppskattning för Hexagon v4 DSP genom exekvering av kod producerad av LLVM och Unison på en simuleringsplattform. Resultaten visar att upskattningen är oriktig, men att det är en stark korrelation mellan en given uppskattning och observerad förbättring. Den största orsaken till felaktigheter i uppskattningen beror på uppehåll i hårdvaran, vilket kod från Unison lider mer av jämfört med kod från LLVM.

Preterm infants are sensitive to incorrect levels of oxygen in the blood. Thus, the oxygen levels need to be monitored continuously. The most commonly used method today has several limitations and therefore constant blood tests has to be taken. Neosense Technologies AB has developed an electrochemical sensor that measures blood oxygen tension continuously and in real time. This thesis aims to develop a prototype algorithm to derive cardiac output based on dynamic changes in inhaled fraction of oxygen and the corresponding change in arterial oxygen tension using the electrochemical sensor. Two measurements were done using the sensor to measure the step response of partial pressure of oxygen to obtain data to develop the algorithm. Using the obtained data, two algorithms were developed and from the validation analysis one of the algorithms were chosen due to significant better results. / För tidigt födda barn är känsliga för felaktiga syrgasnivåer i blodet. Därför krävs det noggranna och kontinuerliga mätningar. Dagens metoder är begränsade och ständiga blodprover behöver tas.Neosense Technologies AB har utvecklat en sensor som mäter syrehalten i blodet kontinuerligt och i realtid.Syftet med detta examensarete är att utveckla en prototypalgoritm för att härleda hjärtminutvolymen baserat på dynamiska förändringar i syrehalten i inandningsluften och den motsvarande syrgasnivån i blodet med hjälp av den elektrokemiska sensorn.Två mätningar gjordes med hjälp av sensorn där stegsvaret för partialtrycket av syre mättes för att erhålla data för algoritmutvecklingen. Med hjälp av den erhållna datan utvecklades två algoritmer och från valideringsanalysen valdes en av algoritmerna på grund av betydande bättre resultat.

This thesis discusses the development of an advanced real-time embedded systems course offered at California Polytechnic State University, San Luis Obispo, which aims to prepare students to design modern complex real-time embedded systems. It describes the goals of the real-time embedded systems curriculum, which includes an introductory and advanced course. Finally, this paper discusses the challenges of creating a successful advanced real-time embedded systems course and proposes changes to the current advanced real-time embedded systems course in response to those challenges.

My thesis explores the effectiveness of software techniques that bend digital abstractions in order to allow embedded systems to do more with less energy. Recent years have witnessed a proliferation of low-power embedded devices with power ranges of few milliwatts to microwatts. The capabilities and size of the embedded systems continue to improve dramatically; however, improvements in battery density and energy harvesting have failed to mimic a Moore's law. Thus, energy remains a formidable bottleneck for low-power embedded systems. Instead of trying to create hardware with ideal energy proportionality, my dissertation evaluates how to use unconventional and probabilistic computing that bends traditional abstractions and interfaces in order to reduce energy consumption while protecting program semantics. My thesis considers four methods that unleash energy otherwise squandered on communication, storage, time keeping, or sensing: 1) CCCP, which provides an energy-efficient storage alternative to local non-volatile storage by relying on cryptographic backscatter radio communication, 2) Half-Wits, which reduces energy consumption by 30% by allowing operation of embedded systems at below-spec supply voltages and implementing NOR flash memory error recovery in firmware rather than strictly in hardware, 3) TARDIS, which exploits the decay properties of SRAM to estimate the duration of a power failure ranging from seconds to several hours depending on hardware parameters, and 4) Nonsensors, which allow operation of analog to digital converters at low voltages without any hardware modifications to the existing circuitry.

Rockfall, which is the detachment of rocks from a mountain, is a major hazard in the mining industry. To help combat this issue, this thesis aims to develop a sensor platform that is able to detect both the potential risk for rockfall and if any rocks do hit the ground. The platform requires wireless communication in order to output relevant information and in order to be part of an IoT-network of sensors.The design of the platform used three different sensors, a geophone, an accelerometer and a microphone. The main focus of the design process was to keep the platform low power, enabling long operation times. The final design had data output from both the microphone and accelerometer, with the accelerometer able to pick up the impact from a falling barbell. Wireless transmission of data is possible for up to 7.5 meters using Bluetooth Low Energy. The low power design was met, with an average current consumption of 26 milliamperes during transmission using Bluetooth Low Energy. That gives a theoretical operation time of 27 days with the battery used. As the accelerometer can pick up a falling object and with a theoretical lifetime of 27 days for the platform, it can be argued that the goals, except for detecting potential risk for rockfall, were met. In order to properly function, however, the platform needs more development, but the most important conclusion of the work is that it seems possible to build this type of platform. Further research and development outside the scope of the thesis is connecting several platforms together.

No description available.