Browsing by Subject "Internet of Things"

Sort by: Order: Results:

Now showing items 1-11 of 11
  • Kumar, Arjun; Hoque, Mohammad Ashraful; Nurmi, Petteri; Pecht, Michael G.; Tarkoma, Sasu; Song, Junehwa (ACM, 2020)
    Deployments of battery-powered IoT devices have become ubiquitous, monitoring everything from environmental conditions in smart cities to wildlife movements in remote areas. How to manage the life-cycle of sensors in such large-scale deployments is currently an open issue. Indeed, most deployments let sensors operate until they fail and fix or replace the sensors post-hoc. In this paper, we contribute by developing a new approach for facilitating the life-cycle management of large-scale sensor deployments through online estimation of battery health. Our approach relies on so-called V-edge dynamics which capture and characterize instantaneous voltage drops. Experiments carried out on a dataset of battery discharge measurements demonstrate that our approach is capable of estimating battery health with up to 80% accuracy, depending on the characteristics of the devices and the processing load they undergo. Our method is particularly well-suited for the sensor devices, operating dedicated tasks, that have constant discharge during their operation.
  • University of Helsinki, Department of Computer Science; Münch, Jürgen; ; (University of Helsinki, Department of Computer Science, 2014)
    There is a need in many software-based companies to evolve their software development practices towards continuous integration and continuous deployment. This allows a company to frequently and rapidly integrate and deploy their work and in consequence also opens opportunities for getting feedback from customers on a regular basis. Ideally, this feedback is used to support design decisions early in the development process, e.g., to determine which features should be maintained over time and which features should be skipped. In more general terms, the entire R&D system of an organization should be in a state where it is able to respond and act quickly based in instant customer feedback and where actual deployment of software functionality is seen as a way of fast experimenting and testing what the customer needs. Experimentation refers here to fast validation of a business model or more specifically validating a value hypothesis. Reaching such a state of continuous experimentation implies a lot of challenges for organizations. Selected challenges are how to develop the "right" software while developing software "right", how to have an appropriate tool infrastructure in place, how to measure and evaluate customer value, what are appropriate feedback systems, how to improve the velocity of software development, how to increase the business hit rate with new products and features, how to integrate such experiments into the development process, how to link knowledge about value for users or customers to higher-level goals of an organization. These challenges are quite new for many software-based organizations and not sufficiently understood from a software engineering perspective. These proceedings contain selected seminar papers of the student seminar Data- and Value-Driven Software Engineering with Deep Customer Insight that was held at the Department of Computer Science of the University of Helsinki. The seminar was held during the fall semester of 2014 from September 1st to December 8th. Papers in the seminar cover a wide range of topics related to the creation of value in software engineering. An interview of startups shows that emerging companies face a number of key decision points that shape their future. Value has a different meaning in different contexts. Embedded devices can be used to gather data and provide more value to the users through analysis and adaptation to circumstances. In entertainment, metrics can provide content creators the chance to react to user behavior and provide a more meaningful user experience. Value creation needs an active approach to software development from the companies: software engineering processes need to be incorporated with proper mechanisms to find the correct stakeholders, elicit requirements that provide the highest value and successfully implement the necessary changes with short development cycles. When the right building blocks are in place, companies are able to quickly deliver new software and leverage data from their products and services to continuously improve the perceived value of software.
  • H. Khajavi, Siavash; Hossein Motlagh, Naser; Jaribion, Alireza; C. Werner, Liss; Holmström, Jan (2019)
    The concept of a digital twin has been used in some industries where an accurate digital model of the equipment can be used for predictive maintenance. The use of a digital twin for performance is critical, and for capital-intensive equipment such as jet engines it proved to be successful in terms of cost savings and reliability improvements. In this paper, we aim to study the expansion of the digital twin in including building life cycle management and explore the benefits and shortcomings of such implementation. In four rounds of experimentation, more than 25,000 sensor reading instances were collected, analyzed, and utilized to create and test a limited digital twin of an office building facade element. This is performed to point out the method of implementation, highlight the benefits gained from digital twin, and to uncover some of the technical shortcomings of the current Internet of Things systems for this purpose.
  • Kovala, Jarkko (Helsingin yliopisto, 2020)
    Internet of Things (IoT) has the potential to transform many domains of human activity, enabled by the collection of data from the physical world at a massive scale. As the projected growth of IoT data exceeds that of available network capacity, transferring it to centralized cloud data centers is infeasible. Edge computing aims to solve this problem by processing data at the edge of the network, enabling applications with specialized requirements that cloud computing cannot meet. The current market of platforms that support building IoT applications is very fragmented, with offerings available from hundreds of companies with no common architecture. This threatens the realization of IoT's potential: with more interoperability, a new class of applications that combine the collected data and use it in new ways could emerge. In this thesis, promising IoT platforms for edge computing are surveyed. First, an understanding of current challenges in the field is gained through studying the available literature on the topic. Second, IoT edge platforms having the most potential to meet these challenges are chosen and reviewed for their capabilities. Finally, the platforms are compared against each other, with a focus on their potential to meet the challenges learned in the first part. The work shows that AWS IoT for the edge and Microsoft Azure IoT Edge have mature feature sets. However, these platforms are tied to their respective cloud platforms, limiting interoperability and the possibility of switching providers. On the other hand, open source EdgeX Foundry and KubeEdge have the potential for more standardization and interoperability in IoT but are limited in functionality for building practical IoT applications.
  • Poteri, Juho (Helsingin yliopisto, 2020)
    The Internet of Things (IoT) paradigm is seeing rapid adoption across multiple domains—industry, enterprise, agriculture, smart cities, households, only to name a few. IoT applications often require wireless autonomy, thereby placing challenging requirements on communication techniques and power supply methods. Wireless networking using devices with constrained energy, as often is the case in wireless sensor networks (WSN), provokes explicit considerations around the conservation of the supplied power on the one hand and the efficiency of the power drawn and energy used on the other. As radio communications characteristically consume the bulk of all energy in wireless IoT systems, this constrained energy budget combined with aspirations for terminal device lifetime sets requirements for the communications protocols and techniques used. This thesis examines two open architecture low-power wide-area network (LPWAN) standards with mesh networking support, along with their energy consumption profile in the context of power-constrained wireless sensor networks. The introductory section is followed by an overview of IoT and WSN foundations and technologies. The following section describes the IEEE 802.15.4 standard and ecosystem, followed by the Bluetooth LE and Bluetooth Mesh standards. A discussion on these standards' characteristics, behavior, and applicability to power-constrained sensor networks is presented.
  • Taivalsaari, A.; Mikkonen, T. (ACM, 2019)
    The Internet of Things (IoT) enables connected devices that are an integral part of the physical world. The possibility to connect, manage, configure and dynamically reprogram remote devices through local and global cloud environments will open up a broad variety of new use cases, services, applications and device categories, and will enable entirely new product and application ecosystems as well. In this paper we discuss the software architecture options of IoT gateways as a follow-up to our earlier paper that defined a taxonomy of software architectures for IoT devices. We summarize several different software architecture options for IoT gateways. These options have a significant impact on the overall end-to-end architecture and topology of IoT systems, e.g., in determining how much computation can be performed on the edge of the network. Based on our observations and industry experiences we then make predictions on the future of gateway solutions and IoT systems more broadly. © 2019 ACM.
  • Tiekstra, Sanne; Dopico-Parada, Ana; Koivula, Hanna; Lahti, Johanna; Buntinx, Mieke (2021)
    Market implementation of active and intelligent packaging (AIP) technologies specifically for fiber-based food packaging can be hindered by various factors. This paper highlights those from a social, economic, environmental, and legislative point of view, and elaborates upon the following aspects mainly related to interactions among food packaging value chain stakeholders: (i) market drivers that affect developments, (ii) the gap between science and industry, (iii) the gap between legislation and practice, (iv) cooperation between the producing stakeholders within the value chain, and (v) the gap between the industry and consumers. We perceive these as the most influential aspects in successful market implementation at a socioeconomic level. The findings are supported by results from quantitative studies analyzing consumer buying expectations about active and intelligent packaging (value perception of packaging functions, intentions to purchase AIP, and willingness to pay more) executed in 16 European countries. Finally, in this paper, we discuss approaches that could direct future activities in the field towards industrial implementation.
  • Rebeiro-Hargrave, Andrew; Hossein Motlagh, Naser; Varjonen, Samu; Lagerspetz, Eemil; Nurmi, Petteri; Tarkoma, Sasu (IEEE, 2020)
    Air pollution is a contributor to approximately one in every nine deaths annually. To counteract health issues resulting from air pollution, air quality monitoring is being carried out extensively in urban environments. Currently, however, city air quality monitoring stations are expensive to maintain, resulting in sparse coverage. In this paper, we introduce the design and development of the MegaSense Cyber-Physical System (CPS) for spatially distributed IoT-based monitoring of urban air quality. MegaSense is able to produce aggregated, privacy-aware maps and history graphs of collected pollution data. It provides a feedback loop in the form of personal outdoor and indoor air pollution exposure information, allowing citizens to take measures to avoid future exposure. We present a battery-powered, portable low-cost air quality sensor design for sampling PM2.5 and air pollutant gases in different micro-environments. We validate the approach with a use case in Helsinki, deploying MegaSense with citizens carrying low-maintenance portable sensors, and using smart phone exposure apps. We demonstrate daily air pollution exposure profiles and the air pollution hot-spot profile of a district. Our contributions have applications in policy intervention management mechanisms and design of clean air routing and healthier navigation applications to reduce pollution exposure.
  • Chen, Liang; Thombre, Sarang; Järvinen, Kimmo; Lohan, Elena Simona; Alén-Savikko, Anette; Leppäkoski, Helena; Bhuiyan, M. Zahidul H.; Bu-Pasha, Shakila; Ferrara, Giorgia Nunzia; Honkala, Salomon; Lindqvist, Jenna; Ruotsalainen, Laura; Korpisaari, Päivi; Kuusniemi, Heidi (2017)
    Internet of Things (IoT) connects sensing devices to the Internet for the purpose of exchanging information. Location information is one of the most crucial pieces of information required to achieve intelligent and context-aware IoT systems. Recently, positioning and localization functions have been realized in a large amount of IoT systems. However, security and privacy threats related to positioning in IoT have not been sufficiently addressed so far. In this paper, we survey solutions for improving the robustness, security, and privacy of location-based services in IoT systems. First, we provide an in-depth evaluation of the threats and solutions related to both global navigation satellite system (GNSS) and non-GNSS-based solutions. Second, we describe certain cryptographic solutions for security and privacy of positioning and location-based services in IoT. Finally, we discuss the state-of-the-art of policy regulations regarding security of positioning solutions and legal instruments to location data privacy in detail. This survey paper addresses a broad range of security and privacy aspects in IoT-based positioning and localization from both technical and legal points of view and aims to give insight and recommendations for future IoT systems providing more robust, secure, and privacy-preserving location-based services.
  • Sjöman, Heikki; Kalasniemi, Jani; Vartiainen, Matti; Steinert, Martin (2018)
    Prototyping (iterative loops of design-build-test) is a proven method of efficiently developing new products. Developing products not only quickly, but that are also fit for purpose, implies engaging the end users and iterating the technology at hand. However, there is currently little research on how engineering design can approach developing connected devices. The purpose of this paper is to distinguish and discuss design approaches that are suitable for connected devices. Internet of Things devices consist of both the physical products themselves and the data that is coming out of the products, which we define as the external and internal data, respectively. They both can be prototyped separately, but since the data acquired can influence the design of the device and vice versa, we propose to link these two together in the product development process. This issue becomes more apparent when designing networks of sensors, e.g., for complex artificial intelligence (AI) databases. We explain the principle by describing the development of 1Balance through six different prototypes for human balance measurement. Technologically quantifying balance is an underused approach for objectively evaluating the state of a human's performance. The authors have developed a mobile application for monitoring balance as a physiological signal (amount of sway) via a compact wireless inertial measurement unit (IMU) sensor strapped to the body of the subject for the duration of the measurement. We describe the design process for developing this connected medical device, as well as how the acquired data was used to improve the design of the product. In conclusion, we propose conceptually connecting the external and internal data prototyping loops.
  • Makitalo, Niko; Flores-Martin, Daniel; Berrocal, Javier; Garcia-Alonso, Jose; Ihantola, Petri; Ometov, Aleksandr; Murillo, Juan Manuel; Mikkonen, Tommi (2020)
    Today, creating innovative Internet of Bodies solutions requires manually gathering the needed information from an increasing number of services and personal devices. In this article, we tackle this challenge by presenting Human Data Model-a programming framework for combining information from several sources, performing computations over that information to high-level abstractions, and then providing these abstractions to proactively schedule computer-human interactions.