Pervasive computing and Internet of Things (IoTs) paradigms have created a huge potential for new business. To fully realize this potential, there is need common way to abstract the heterogeneity devices so that their functionality can be represented as virtual platform. end, we present novel semantic level interoperability architecture pervasive IoTs. There are two main principles in proposed architecture. First, information capabilities with web knowledge representation technologies...

This paper presents a publish-subscribe architecture designed to support information level interoperability in smart space applications the Internet of Things (IoT). The is built on top generic SPARQL endpoint where publishers and subscribers use standard Updates Queries. Notifications about events [i.e., changes resource description framework (RDF) knowledge base] are expressed terms added removed binding results since previous notification, limiting network overhead facilitating...

Energy markets are currently evolving due to the rapid digitalization, increasing amount of renewable energy sources, and technology development. Market actors being forced adapt their business models these changes or find new opportunities. In our earlier work, change scenarios were identified that reflect future potential Finnish companies in domain particularly with respect demand flexibility, actor roles, consumer services novel activities a kind market. The purpose this paper is examine...

In this paper we present fully automatic interpatient electrocardiogram (ECG) signal classification method using deep convolutional neural networks (CNN). ECG is simple and non-invasive way to measure the electric activity of heart it provides valuable information about condition heart. It widely utilized for detecting different abnormalities in heartbeat. Identifying time-consuming, because often requires analyzing each heartbeat recording. Therefore, arrhythmias machine-learning...

Semantics associates meaning with Internet of Things (IoT) data and facilitates the development intelligent IoT applications services. However, big volume generated by devices resource limitations these have given rise to challenges for applying semantic technologies. In this article, we present Cloud edge based architectures reasoning. We report three experiments that demonstrate how computing can facilitate systems in terms transfer also analyze distributing reasoning tasks between affects...

It has been proposed that Semantic Web technologies would be key enablers in achieving context-aware computing our everyday environments. In vision of semantic technology empowered smart spaces, the whole interaction model is based on sharing data via common blackboards. This approach allows space applications to take full advantage technologies. Because its novelty, there is, however, a lack solutions and methods for developing according this vision. paper, we present most relevant...

Accurate short-term forecasts of building energy consumption are necessary for profitable demand response. Short-term forecasting methods can be roughly classified into physics-based modelling and data-based modelling. Both these approaches have their advantages disadvantages it would therefore ideal to combine them. This paper proposes a novel approach that allows us the best parts machine learning while avoiding many drawbacks. A key idea in is provide variety parameters as input an...

In this paper we present a novel knowledge sharing protocol (KSP) for semantic technology empowered ubiquitous computing systems. particular the is designed M3 which blackboard based interoperability solution smart spaces. The main difference between KSP and existing work that provides SPARQL-like mechanisms in compact binary format to be suitable also resource restricted devices networks. order evaluate practice implemented case study prototype space, called Smart Greenhouse. messages were...

Demand-side flexibility management is a key enabler of the transformation towards high penetration renewable energy resources. We present flexibility-management system called Flex4Grid, which designed to provide low-cost solution for residential consumers wishing participate in power-grid balancing. The Flex4Grid continuously forecasts need power grid and informs about periods. Consumers can their an aggregator exchange reward, depends on selected incentive scheme. automation events provided...

The current approaches for energy consumption optimisation in buildings are mainly reactive or focus on scheduling of daily/weekly operation modes heating. Machine Learning (ML)-based advanced control methods have been demonstrated to improve efficiency when compared these traditional methods. However, placing ML-based models close the is not straightforward. Firstly, edge-devices typically lower capabilities terms processing power, memory, and storage, which may limit execution inference at...

This paper presents a novel deep learning architecture for short-term load forecasting of building energy loads. The is based on simple base learner and multiple boosting systems that are modelled as single neural network. transforms the original multivariate time series into cascading univariate series. Together with sparse interactions, parameter sharing equivariant representations, this approach makes it possible to combat against overfitting while still achieving good presentation power...

Data are crucial to support decision making. If data have low veracity, decisions not likely be sound. Internet of Things (IoT) generates big with inaccuracy, inconsistency, incompleteness, deception, and model approximation. Enhancing veracity is important address these challenges. In this article, we summarize the key characteristics challenges IoT, which influence processing We review landscape measuring enhancing mining uncertain streams. Moreover, propose five recommendations for future...

Optimal management of demand-side flexibility in buildings is important for properly integrating large amounts intermittent generation from windmills and photovoltaics. This paper proposes a novel Energy Management Agent (EMA) concept that can optimize building's energy costs with respect to external prices while at the same time allow be used via explicit demand response. The EMA combines Artificial Neural Networks (ANN) model predictive control modelling optimization flexibility. It...

Smart Spaces are an emerging computing paradigm in which various devices of our everyday environment share information autonomously with each other order to provide services people the environment. Achieving reliable way between heterogeneous is not easy process however. Solutions that independent application domain and interoperability for Space needed. This paper presents a new approach low capacity embedded systems be part applications use cases based on semantic interoperability. The...

Enabling smart applications to access information about the physical world in a machine interpretable format is high priority Internet of Things (IoT) related research. In this paper, we present novel approach for advertising objects user vicinity. There are three distinctive features our approach. First, ubiquitous codes (ucodes) used providing globally unique identifiers objects. Second, use Bluetooth Low Energy beacons broadcasting Third, object represented with semantic technologies....

Our everyday living environments are inhabited by a huge and constantly increasing number of electronic devices. Smart Environment is common name for physical place where these various devices interact with each other in order to provide useful services the end-user. Because different people usually have kind needs it important that user able configure so provided individual fact useful. In this paper, we present an approach enabling end-user how Environments functions. our utilize Semantic...

In this paper we evaluate two M3 smart space information broker implementations -- Smart-M3 and RIBS. is a semantic interoperability architecture for both local Internet connected environments it utilizes RDF databases in similar fashion as does the Semantic Web. The compared have functionality, but their design philosophies are very different. an open source reference RIBS solution devices with restricted computational capabilities. Our evaluation based on qualitative quantitative...

Support of legacy devices and services is crucial for the adoption new smart space technologies. We present two technologies which enable formation local ad-hoc spaces with commercial off-the-shelf (COTS) devices. First these NoTA, a service oriented architecture enabling networks different physical transports. Second one Smart-M3, semantic information sharing spaces. It aims at opening world use applications in world, thus types mash-up applications. In our demonstration - Open-M3 we show...

The ability to shift energy consumption a time when lot of is available supports the integration renewable sources. Services that intend do this automatically are currently emerging in markets. These kinds flexibility services targeted as smart household consumers, promising them cost savings, optimal living conditions, and other benefits. purpose paper examine current demand offering these Finland. We performed an extensive consumer survey untangle consumers' potential, motivation,...