Embedded devices constantly face two challenges in data security and energy efficiency. These devices are limited in processing such secure functions, as well as maintaining enough energy for the device to function properly. One example involves the healthcare industry, where some patients may require an Implantable Cardioverter Defibrillator (ICD) in their hearts to measure the heartbeat rate, while powered by a battery. The heartbeat rate is sent wirelessly, and the ICD can receive a jolt of electricity when the heartbeat rate reaches an abnormal value. Transmitting data alone, however, yields potential security risks when sending plain data. Work has shown that an attacker could intercept the heartbeat rate of the ICD, and intentionally send jolts of electricity. Also, replacing the battery on an ICD involves quite a painful process for the patient. A battery-less device that can receive energy wirelessly is much more convenient, but also poses a challenge where power loss may occur under long distances due to a limited supply of energy. In this paper, we design an adaptive light-weight scheduling mechanism that enhances data security, as well as improving energy efficiency on a device with such constraints. We will then prototype this scheduler on a Wireless Identification and Sensing Platform (WISP) device, which includes these constraints. Our results will then demonstrate the capabilities of such adaptive scheduling under various distances.

Demand for compact, easily deployable, energy-efficient computers has driven the development of general-purpose transiently powered computers (TPCs) that lack both batteries and wired power, operating exclusively on energy harvested from their surroundings. TPCs' dependence solely on transient, harvested power offers several important design-time benefits. For example, omitting batteries saves board space and weight while obviating the need to make devices physically accessible for maintenance. However, transient power may provide an unpredictable supply of energy that makes operation difficult. A predictable energy supply is a key abstraction underlying most electronic designs. TPCs discard this abstraction in favor of opportunistic computation that takes advantage of available resources. A crucial question is how should a software-controlled computing device operate if it depends completely on external entities for power and other resources? The question poses challenges for computation, communication, storage, and other aspects of TPC design. The main idea of this work is that software techniques can make energy harvesting a practicable form of power supply for electronic devices. Its overarching goal is to facilitate the design and operation of usable TPCs. This thesis poses a set of challenges that are fundamental to TPCs, then pairs these challenges with approaches that use software techniques to address them. To address the challenge of computing steadily on harvested power, it describes Mementos, an energy-aware state-checkpointing system for TPCs. To address the dependence of opportunistic RF-harvesting TPCs on potentially untrustworthy RFID readers, it describes CCCP, a protocol and system for safely outsourcing data storage to RFID readers that may attempt to tamper with data. Additionally, it describes a simulator that facilitates experimentation with the TPC model, and a prototype computational RFID that implements the TPC model. To show that TPCs can improve existing electronic devices, this thesis describes applications of TPCs to implantable medical devices (IMDs), a challenging design space in which some battery-constrained devices completely lack protection against radio-based attacks. TPCs can provide security and privacy benefits to IMDs by, for instance, cryptographically authenticating other devices that want to communicate with the IMD before allowing the IMD to use any of its battery power. This thesis describes a simplified IMD that lacks its own radio, saving precious battery energy and therefore size. The simplified IMD instead depends on an RFID-scale TPC for all of its communication functions. TPCs are a natural area of exploration for future electronic design, given the parallel trends of energy harvesting and miniaturization. This work aims to establish and evaluate basic principles by which TPCs can operate,

Embedded devices in the Internet of Things world mostly only have one connection channel, and smaller consumer devices usually communicate with other devices only over a wireless connection. Developers constantly upgrade the Internet of Things devices with Android phones connecting to the devices. If developers break the only possible connection, the devices become unusable, because the devices do not have a wired port. This study explores the options to test the wireless upgrade channel during the development workflow and implements a continuous integration pipeline for the devices. Literature in the field of continuous integration focuses primarily on Cloud and Web-related workloads. The few papers targeting embedded devices with continuous integration are primarily theoretical and discuss the possible advantages of utilizing continuous integration but do not implement a prototype. Our contribution creates a continuous integration pipeline for embedded devices, which automatically tests the update channel between Android phones and embedded devices over Bluetooth on every commit. To verify the correct functionality, we use previously faulty commits from the git history and run the pipeline with the buggy commit to check the pipeline’s functionality. The pipeline evaluation shows that the time improvements for our update validation process with continuous integration is insignificantly faster for the upgrade procedure. However, developers are required to test a combination of versions and here, the automated testing setup excels over the human testing method by being scalable. Furthermore, automated testing enables easier identification of the root cause for an issue and a faster delivery time of fixes. While the pipeline works reliably, we identify issues in adopting the continuous integration process by the developers. Additionally, the analysis summarizes essential tools and features to run the pipeline with an overview of required elements for similar projects. The work was created in cooperation with Wrlds Creations AB and we used the devices actively developed by the company. / Inbäddade enheter i Internet of Things världen har oftast bara en kommunikationskanal, och mindre konsumentenheter kommunicerar vanligtvis via en trådlös Bluetooth-anslutning. Utvecklarna uppgraderar ständigt enheterna med Android-telefoner som ansluts till enheterna. Om utvecklarna bryter den enda möjliga kommunikationskanalen blir enheterna obrukbara. Denna studien undersöks alternativen för att testa den trådlösa uppgraderingskanalen under utvecklingsarbetsflödet och implementeras en continuous integration pipeline för enheterna. Litteraturen inom området continuous integration fokuserar primärt på Cloud och webbrelaterade arbetsbelastningar. De få artiklar som är inriktade på inbäddade enheter med continuous integration är främst teoretiska och diskuterar de möjliga fördelarna med att använda continuous integration, men implementerar inte någon prototyp. Vårt bidrag skapar en continuous integration pipeline för inbäddade enheter, som automatiskt testar uppdateringskanalen mellan Android-telefoner och inbäddade enheter via Bluetooth vid varje commit. För att verifiera den korrekta funktionaliteten använder vi tidigare felaktiga commits från git-historiken och körde pipelinen med den felaktiga commit för att kontrollera funktionaliteten av pipelinen. Utvärderingen av pipelinen visar att tidsförbättringarna för valideringskanalen är obetydligt snabbare för hela uppgraderingsproceduren. Utvecklarna krävs dock testa en kombination av versioner och här är den automatiska testuppställningen bättre än den mänskliga testmetoden eftersom den är skalbar. Automatiserad testning gör det dessutom lättare att identifiera grundorsaken till ett problem och att leverera korrigeringar snabbare. Även om pipelinen fungerar på ett tillförlitligt sätt identifierar vi problem när det gäller utvecklarnas antagande av continuous integration-processen. Dessutom sammanfattas i analysen de viktigaste verktygen och funktionerna för att driva pipelinen med en översikt över vad som krävs för liknande projekt. Arbetet skapades i samarbete med Wrlds Creations AB och använde de enheter som företaget aktivt utvecklat.

Embedded systems are ubiquitous in our society, they control vehicles, aircrafts and medical instruments. Some of these systems are distributed, which means they are part of a network and their operation is coordinated. Software development for such systems can be a difficult problem. In this thesis we propose SOFA 2 component system to simplify the software development for distributed embedded systems where the distribution of components is handled entirely by the component system. Lego Mindstorms is proposed as the model of a distributed embedded system. A runtime environment for SOFA 2 and a demo application were developed to evaluate the approach. The proposed approach delivers seamless component distribution. Nevertheless, non-functional requirements such as memory, program size or disk space must be included in the implementation to fully benefit from a component system.

From the last decade, studies on human facial emotion recognition revealed that computing models based on regression modelling can produce applicable performance. In this study, an automatic facial expression real-time system was built and tested. The method is used in this study has been used widely in different areas such as Local Binary Pattern method, which has been used in many research projects in machine vision, and the K-Nearest Neighbour algorithm is method utilized for regression modelling. In this study, these two techniques has been used and implemented on the FPGA for the first time, on the side and joined together to great the model in such way to display a continues and automatic emotional state detection model on the monitor. To evaluate the effectiveness of the classifier technique for human emotion recognition from video, the model was designed and tested on MATLAB environment and then MATLAB Simulink environment that is capable of recognizing continuous facial expression in real time with a rate of 1 frame per second and implemented on a desktop PC. It has been evaluated in a testing dataset and the experimental results were promising with the accuracy of 51.28%. The datasets and labels used in this study are made from videos which, recorded twice from 5 participants while watching a video. In order to implement it in real-time in faster frame rate, the facial expression recognition system was built on FPGA. The model was built on Atlys™ Spartan-6 FPGA Development Board. It can perform continuously emotional state recognition in real time at a frame rate of 30 with the accuracy of 47.44%. A graphic user interface was designed to display the participant video in real time and also two dimensional predict labels of the emotion at the same time. This is the first time that automatic emotional state detection has been successfully implemented on FPGA by using LBP and K-NN techniques in such way to display a continues and automatic emotional state detection model on the monitor.

Home routers are common in every household that has some kind of Internet connectivity. These embedded devices are running services such as web, file and DHCP server. Even though they have the same security issues as regular computers, they do no run protection software such as anti-virus and they are not updated. Moreover, the importance of these devices is misjudged; all network traffic is passing through them and they control the DNS of the network while, in most cases, they are on-line around the clock. When more and more non-Internet features are implemented into home routers, such as Voice over IP and network storage, their role becomes more special and many security concerns are raising. In this thesis, we investigate the issues resulting from this special role; the importance for these devices to be secure, the attacking vector and how the devices can be compromised to be part of a large home router botnet. We conclude by proposing ways to make the current implementation more secure, suggesting ways to protect routers from botnets without user interaction, that is from the ISP, while respecting the privacy of the end user and we identify what future work needs to be done. / Router är vanliga i hem som har någon slags Internet anslutning. De här inbyggda enheter kör tjänster som t.ex. web, file och DHCP basenheter. Fastän de har samma säkerhetsfrågor som vanliga datorer, så kan de inte använda säkerhets mjukvara som t.ex anti-virus och de är inte uppdaterade. Dessutom har betydelsen av de här apparaterna blivit felbedömmat; hela nätverket passerar genom dem och de kontrolerar nätverkets DNS medan, i de flesta fall, de är on-line dygnet runt. Men, när mer och mer icke-Internet lockvaror fars in i routern, som t.ex Voice över IP och nätverkslagring, blir deras roll viktigare och oron för säkerheten växer. I den här avhandlingen utforskars problemen och frågorna som efterföljer deras speciella roll, hur viktigt det är att de här apparaterna är skyddade, (the attacking vector) och hur de här apparaterna kan bli jämkningad för att bli en del av ett stort router botnet. Vi avsluter med att lägga fram sätt att göra det nuvarande verktyget mer skyddat, föreslå sätt att skydda routern från botnet utan användarinteraktion, som kommer från ISP, medan man respekterar det andra användarens privtaliv och markera vad som behövs ändras i framtiden.

<p>Deep Neural Networks (DNNs) are a class of machine learning algorithms that are widelysuccessful in various computer vision tasks. DNNs filter input images and videos with manyconvolution operations in each layer to extract high-quality features and achieve high ac-curacy. Although highly accurate, the state-of-the-art DNNs usually require server-gradeGPUs, and are too energy, computation and memory-intensive to be deployed on most de-vices. This is a significant problem because billions of mobile and embedded devices that donot contain GPUs are now equipped with high definition cameras. Running DNNs locallyon these devices enables applications such as emergency response and safety monitoring,because data cannot always be offloaded to the Cloud due to latency, privacy, or networkbandwidth constraints.</p> <p>Prior research has shown that a considerable number of a DNN’s memory accesses andcomputation are redundant when performing computer vision tasks. Eliminating these re-dundancies will enable faster and more efficient DNN inference on low-power embedded de-vices. To reduce these redundancies and thereby reduce the energy consumption of DNNs,this thesis proposes a novel Tree-based Unidirectional Neural Network (TRUNK) architec-ture. Instead of a single large DNN, multiple small DNNs in the form of a tree work togetherto perform computer vision tasks. The TRUNK architecture first finds thesimilaritybe-tween different object categories. Similar object categories are grouped intoclusters. Similarclusters are then grouped into a hierarchy, creating a tree. The small DNNs at every nodeof TRUNK classify between different clusters. During inference, for an input image, oncea DNN selects a cluster, another DNN further classifies among the children of the cluster(sub-clusters). The DNNs associated with other clusters are not used during the inferenceof that image. By doing so, only a small subset of the DNNs are used during inference,thus reducing redundant operations, memory accesses, and energy consumption. Since eachintermediate classification reduces the search space of possible object categories in the image,the small efficient DNNs still achieve high accuracy.</p> <p>In this thesis, we identify the computer vision applications and scenarios that are wellsuited for the TRUNK architecture. We develop methods to use TRUNK to improve the efficiency of the image classification, object counting, and object re-identification problems.We also present methods to adapt the TRUNK structure for different embedded/edge ap-plication contexts with different system architectures, accuracy requirements, and hardware constraints.</p> <p>Experiments with TRUNK using several image datasets reveal the effectiveness of theproposed solution to reduce memory requirement by∼50%, inference time by∼65%, energyconsumption by∼65%, and the number of operations by∼45% when compared with existingDNN architectures. These experiments are conducted on consumer-grade embedded systems:NVIDIA Jetson Nano, Raspberry Pi 3, and Raspberry Pi Zero. The TRUNK architecturehas only marginal losses in accuracy when compared with the state-of-the-art DNNs.</p>

No description available.

Hand pose estimation is a computer vision challenge that consists of detecting the coordinates of a hand’s key points in an image. This research investigates several deep learning-based solutions to determine whether or not it is possible to improve current state-of-the-art detectors for smartphone applications. Several models are tested and compared based on accuracy, processing speed and memory size. A final network is selected and detailed to compare it to the state-of-the-art. The proposed solution is obtained by combining the Differentiable Spatial to Numerical Transform layer to predict numerical coordinates together with the Fire module presented in the SqueezeNet architecture. This deep neural network contains around 1 million parameters and is able to outperform the current best documented model in all the metrics described above. A qualitative analysis is also performed to examine the predictions of the final solution on test images. / Att bestämma en hands orientering är en utmaning inom bildanalys som består i att detektera koordinaterna för olika nyckelpunkter för handen i en bild. I denna studie undersöks ett antal metoder baserade på djupinlärning för att avgöra huruvida det är möjligt att förbättra existerande detektorer för tillämpningar på smartphones. Flera olika modeller testas och jämförs baserat på noggrannhet, beräkningshastighet och minneskrav. Ett slutligt nätverk väljs, analyseras och jämföras med nuvarande state-of-the-art teknik. Den lösning som föreslås erhålls genom att kombinera ett så kallat Differentiable Spatial to Numerical Transform-lager, för att förutsäga numeriska koordinater, tillsammans med en så kallad Fire-modul som tidigare presenteras som en del av arkitekturen SqueezeNet. Detta djupa neurala nätverk innehåller cirka en miljon parametrar och kan överträffa den nuvarande mest dokumenterade modellen i alla de avseenden som beskrivits ovan. En kvalitativ analys utförs också för att undersöka den slutliga lösningens uppskattningar på testbilder.

Graphics Processing Units (GPUs) are commonly used to accelerate massively parallel workloads across a wide range of applications from machine learning to cryptocurrency mining. The original application for GPUs, however, was to accelerate graphics rendering which remains popular today through video gaming and video rendering. While GPUs began as fixed function hardware with minimal programmability, modern GPUs have adopted a design with many programmable cores and supporting fixed function hardware for rasterization, texture sampling, and render output tasks. This balance enables GPUs to be used for general purpose computing and still remain adept at graphics rendering. Previous work at the Georgia Institute of Technology has been done to implement a general purpose GPU (GPGPU) in the open source RISC-V ISA. The implementation features many programmable cores and texture sampling support. However, creating a truly modern GPU based on the RISC-V ISA requires the addition of fixed function hardware units for rasterization and render output tasks in order to meet the demands of current graphics APIs such as OpenGL or Vulkan. This thesis discusses the work done by students at the Georgia Institute of Technology and California Polytechnic State University SLO to accelerate graphics rendering on RISC-V GPUs including the specific contributions made to implement and connect fixed function graphics hardware for the render output unit (ROP) to the programmable cores in a RISC-V GPU. This thesis also explores the performance and area cost of different hardware configurations within the implemented GPU.