Smart home technology and IoT is rapidly gaining popularity, and many systems use mobile applications to control the hardware in the homes of the owners. Connective created by Indentive is one such system, which allows users to control things like the temperature in their homes via their smartphones. However, the usability of applications such as this one can be limited, which leads to frustration and even loss of interest for the users. This can result in critical implications for the company producing such a system – both economically and in regards to popularity. It is therefore important for companies to detect issues with their systems early on in order to address these issues and prevent negative implications. Connective was thus tested in a qualitative study using contextual inquiries with 7 participants. The data from the inquiries was then structured through thematic analysis, and the usability of the application was also tested using SUS-evaluation. Wireframe suggestions for changes to the application were then brainstormed with the thematic analysis and the result of the SUS in mind. Major issues that were addressed were access to help, information, the layout, the handling of errors, and feedback. Future studies should include further, more in-depth qualitative studies, perhaps in combination with quantitative data, to further improve and develop the wireframe-suggestions into working prototypes that can then be user tested.

The entry of information technology into our homes and workplaces has changed our way of life. Technologies such as the Internet of Things and Artificial Intelligence are increasingly being used by companies today as means to create new smart services. Digital transformation is happening here and now across many industries. A digital transformation involves a major change in the company and its business model with the aim of creating value. Due to the extensive nature of the digital transformation, there are good incentives to thoroughly investigate existing barriers before a change work can begin. Many studies have been conducted aimed at examining barriers to digital development in, for example, the production industry, but few studies have been conducted aimed at examining barriers within the service sector. This study aims to investigate the barriers that can hamper a digital transformation of an insurance company. In order to answer this research question, a case study was conducted with five semi-structured interviews with informants at an insurance company. The results show that several different barriers exist, most of which can be linked to technical, environmental, organizational and cultural barriers.

Les réseaux centrés contenus (ICN) sont considérés comme une solution aux nouveaux défis et modes de communication liés à l'émergence de l'Internet des Objets (IoT). Pour confirmer cette hypothèse, la problématique fondamentale du routage sur les réseaux ICN-IoT doit être abordée. Cette thèse traite de ce sujet à travers l'architecture IoT.Premièrement, une méthode sécurisée est introduite pour acheminer des paquets ICN à partir de coordonnées géographiques dans un réseau sans-fil de capteurs à faible puissance. Elle est comparée à une inondation optimisée du réseau inspirée des approches existant dans la littérature. En particulier, leur faisabilité et passage à l'échelle sont évalués via un modèle mathématique. Le modèle est paramétré grâce à des données réalistes issues de simulation, de la littérature, et d'expériences sur des capteurs. Il est montré que le routage géographique permet de diviser la mémoire nécessaire sur les capteurs par deux et de réduire considérablement le coût énergétique du routage, en particulier pour des topologies dynamiques.Ensuite, ICN est utilisé pour contrôler l'admission à une plate-forme de calcul de type Fog afin de garantir le temps de réponse. La stratégie de contrôle d'admission proposée, le LRU-AC, utilise l'algorithme Least-Recently-Used (LRU) pour apprendre en direct la distribution de popularité des requêtes. Son efficacité est démontrée grâce à un modèle fondé sur un réseau de files d'attente. Une implémentation du LRU-AC est proposé, utilisant des filtres de Bloom pour satisfaire aux contraintes des cartes FPGA. Son bien-fondé est prouvé par un modèle mathématique et son efficacité en termes de latence et débit démontrée.Enfin, on présente vICN, un outil pour la gestion et la virtualisation de réseaux ICN-IoT. Il s'agit d'une plate-forme qui unifie la configuration et la gestion des réseaux et des applications en exploitant les progrès des techniques d'isolation et de virtualisation. vICN est flexible, passe à l'échelle, et peut remplir différents buts : expériences à grande échelle reproductibles pour la recherche, démonstrations mélangeant machines émulées et physiques, et déploiements réels des technologies ICN dans les réseaux IP existants. / As the Internet of Things (IoT) has brought upon new communication patterns and challenges, Information-Centric Networking (ICN) has been touted as a potential solution. To confirm that hypothesis, the fundamental issue of routing and forwarding in the ICN-IoT must be addressed. This thesis investigates this topic across the IoT architecture.First, a scheme to securely forward ICN interests packets based on geographic coordinates is proposed for low-power wireless sensor networks (WSN). Its efficiency is compared to an optimized flooding-based scheme similar to current ICN-WSN approaches in terms of deployability and scalability using an analytical model. Realistic data for the model is derived from a mixture of simulation, literature study, and experiments on state-of-the-art sensor boards. Geographic forwarding is shown to halve the memory footprint of the ICN stack on reference deployments and to yield significant energy savings, especially for dynamic topologies. Second, ICN is used to enhance admission control (AC) to fixed-capacity Edge-computing platforms to guarantee request-completion time for latency-constrained applications. The LRU-AC, a request-aware AC strategy based on online learning of the request popularity distribution through a Least-Recently-Used (LRU) filter, is proposed. Using a queueing model, the LRU-AC is shown to decrease the number of requests that must be offloaded to the Cloud. An implementation of the LRU-AC on FPGA hardware is then proposed, using Ageing Bloom Filters (ABF) to provide a compact memory representation. The validity of using ABFs for the LRU-AC is proven through analytical modelling. The implementation provides high throughput and low latency.Finally, the management and virtualization of ICN-IoT networks are considered.vICN (virtualized ICN), a unified intent-based framework for network configuration and management that uses recent progress in resource isolation and virtualization techniques is introduced. It offers a single, flexible and scalable platform to serve different purposes, ranging from reproducible large-scale research experimentation to demonstrations with emulated and/or physical devices and network resources and to real deployments of ICN in existing IP networks.

This thesis investigates the possibility of porting a neural network model trained and modeled in TensorFlow to a low-power AI inference accelerator for IoT edge computing. A slightly modified LeNet-5 neural network model is presented and implemented such that an input frequency of 10 frames per second is possible while consuming 4mW of power. The system is simulated in software and synthesized using the FreePDK45 technology library. The simulation result shows no loss of accuracy, but the synthesis results do not show the same positive results for the area and power. The default version of the accelerator uses single-precision floating-point format, float32, while a modified accelerator using the bfloat16 number representation shows significant improvements in area and power with almost no additional loss of accuracy.

No description available.

This thesis focuses on Internet of Things (IoT) and open-source technologies based on it. Specifically aims at software solutions relevant to smart home and compatible with Raspberry Pi platform and MQTT communication protocol. This thesis also focuses on studying Petri Net Virtual Machine (PNVM) and its potential application in IoT. The objective is to design integration of PNVM into IoT with help of existing software means compatible with MQTT and implement a demo application for smart home.

The subject of this diploma thesis, is the Internet of Things, and its use in applied practice, with a subsequent design of a custom system of smart home automatization. It acquaints the reader with the issues of automated home and smart systems, including a basic familiarization of regulations as declared by law about electronic devices. The attributes of miniature computer platforms, developmental boards, and wireless modules for the design of custom systems are analyzed and compared. The biggest part of the thesis is dedicated to a design of a custom system, specifically designing the central processing module and other controlling modules for controlling various electric home appliances. The main controlling unit of the system is a Raspberry Pi 2B unit, with which it is possible to control a device in a household through an user-friendly web application.

The diploma thesis deals with the implementation of own design of an electronic security system with the possibility of extension with the elements of home automation. The control panel is built using the Raspberry Pi3 B+ with a touch screen that communicates with the sensors using Bluetooth Low Energy technology.

This diploma thesis consists of theoretical description of M2M communication and main differences comparing with H2H communication. There is also a description of protocols used in Internet of Things and description of practical implementation of protocols WM-BUS and Z-Wave in Python programming language with usage of Django framework.

The goal of this thesis is to create control software for stand-alone payment machines. The purpose of this software is minimize maintenance cost and usability as much as possible. Description of technologies and protocols used in implementation are in the text. Next, there is process of work beginning with analysis, continues with design, construction, testing and installation manual. Projekt status and product utilization at the end of project is in chapter about evaluation.