Search

Decision support systems are becoming increasingly sophisticated (e.g., being machine learning-based), attempting to automate decisions as much as possible. However, it remains challenging to extract meaningful value from large quantities of data while also maintaining transparency in seeking justification for the choices made. Instead of creating methods for increasing the interpretability of black box models, one way forward is to design models that are inherently interpretable in the first place. Rule-based methods can automate decisions with great transparency and accuracy, helping to ensure compliance with regulations and adherence to organizational guidelines. In this paper, we propose an approach that uses a decision tree machine learning classification technique for extracting business rules from IoT-generated data to predict the asset status of Smart Returnable Transport Items (SRTIs). We report on an industrial case study that uses two years of historical data, obtained from an SRTI provider in the Netherlands, to predict the status of smart pallets. We compare the performance with the results obtained by using a support-vector machine (SVM) technique. Our experiments show that our solution is both accurate and flexible in terms of business rule elicitation. The obtained decision trees are human-interpretable, can easily be combined with other decision-making techniques, and provide a prediction accuracy marginally higher than an SVM technique.

Detection of fatigue helps prevent injuries and optimize the performance of horses. Previous studies tried to determine fatigue using physiological parameters. However, measuring the physiological parameters, e.g., plasma lactate, is invasive and can be affected by different factors. In addition, the measurement cannot be done automatically and requires a veterinarian for sample collection. This study investigated the possibility of detecting fatigue non-invasively using a minimum number of body-mounted inertial sensors. Using the inertial sensors, sixty sport horses were measured during walk and trot before and after high and low-intensity exercises. Then, biomechanical features were extracted from the output signals. A number of features were assigned as important fatigue indicators using neighborhood component analysis. Based on the fatigue indicators, machine learning models were developed for classifying strides to non-fatigue and fatigue. As an outcome, this study confirmed that biomechanical features can indicate fatigue in horses, such as stance duration, swing duration, and limb range of motion. The fatigue classification model resulted in high accuracy during both walk and trot. In conclusion, fatigue can be detected during exercise by using the output of body-mounted inertial sensors.

In this paper, we propose Anchor-agnostic Transformers (AaTs) that can exploit the attention mechanism for Received Signal Strength (RSS) based fingerprinting localization. In real-world applications, the RSS modality is inherently well-known for its extreme sensitivity to dynamic environments. Since most machine learning algorithms applied to the RSS modality do not possess any attention mechanism, they can only capture superficial representations, yet subtle but distinct ones characterizing specific locations, thereby leading to significant degradation in the testing phase. In contrast, AaTs are enabled to focus exclusively on relevant anchors at every Received Signal Strength (RSS) sequence for these subtle but distinct representations. This also facilitates the model to neglect redundant clues formed by noisy ambient conditions, thus achieving better accuracy in fingerprinting localization. Moreover, explicitly resolving collapse problems at the feature level (i.e., none-informative or homogeneous features) can further invigorate the self-attention process, by which subtle but distinct representations to specific locations are radically captured with ease. To this end, we enhance our proposed model with two sub-constraints, namely covariance and variance losses that are mediated with the main task within the representation learning stage towards a novel multi-task learning manner. To evaluate our AaTs, we compare the models with the state-of-the-art (SoTA) methods on three benchmark indoor localization datasets. The experimental results confirm our hypothesis and show that our proposed models could provide much higher accuracy.

In this paper, it is shown that lower transmission/sampling rates can be used in human activity recognition using channel state information (F1-scores > 85%) and that extremely high sampling rates are unnecessary once the system has been deployed. This is done by analysing the effects of interpolating different sampling rates on Wi-Fi dynamic channel state information for human activity recognition. While current research focuses on training and testing with homogeneous and very high sampling rates (> 100 Hz), this paper outlines some issues with higher sampling rates and explores the impact of training and testing with heterogeneous sampling rates in order to advance more towards joint communication and sensing, where one cannot be certain of the received data rate over time while not knowing the exact training set due to weight sharing in Federated Learning. This paper shows the effect of training and testing with heterogeneous sampling rates (including interpolated datasets) on convolutional neural networks in WiFi sensing.

Unlike motor-vehicle transport, the implementation of lnformation and Communications Technologies (ICT) and Cooperative Intelligent Transport Systems (C-ITS) in cycling has not been comprehensively investigated [1]. Cycling offers several benefits both to society and the environment and is one of the most sustainable and green transportation modes [2]. Many people worldwide have been switching to bicycles during the last decades, and this has increased even more due to the Covid pandemic [3]. Furthermore, the number of people who ride an e-bike is also rising [ 4]. Thus, the number of cyclists is increasing and, in turn, the number of cycling accidents is increasing too. For instance, in the Netherlands, one of the most cycling-friendly countries, 31 % of all road fatalities in 2019 were cyclists (203 fatalities), while in 2020, it was 37% (229 fatalities). One-third of these fatalities were e-bike users [5]. Despite the constantly evolving landscape of cycling and electric bike adoption, applications of new technologies in bicycles are still immature. In recent years, academic research on new technologies related to cyclists' comfort and safety is growing [6, 7, 8]. Furthermore, many studies focus on technologies affecting cyclists' road safety; however, it is unclear what type of technologies are implemented for bicycles. To the best of the authors' knowledge, a comprehensive review of such studies is lacking. Additionally, a clear definition of a 'smart bike'- a concept gaining popularity nowadays, is missing in the literature. To address this gap, the objective of this paper is two fold: 1) to review the state-of-the-art technologies implemented in bicycles to improve cyclists' safety, and 2) to propose an original classification for the levels of smartness of newly emerging 'smart bikes'.

A premature crack is a significant indicator for early failure detection for underground polyvinyl chloride (PVC) water pipes. Using an array of strain gauges mounted on a pipe surface to monitor the strain of adjacent areas of a premature crack is a novel technology that has not been explored. To reduce the risks and upfront investments, we need a testbed to investigate, verify, and validate the innovative technology. However, establishing such a pipe monitoring testbed that covers completely realistic underground situations is challenging. The main reason lies in three main challenges: (i) mimicking the natural changes of water flowing in pipes; (ii) identifying the proper placement of strain gauges and detectable crack sizes; (iii) simulating the crucial underground conditions such as temperature and external stress. To this end, in this paper, we present a testbed to get more insights into the effects of different crack types on the pressure-strain characteristic in realistic conditions. In particular, we use pressure meters and water pumps to control the water flow (for challenge (i)); deploy various strain gauges types and induce cracks with different sizes (for challenge (ii)); fill the pipe with water at various temperatures and underground-like external stress (for challenge (iii)); Analyzing experimental results reveals useful hints for designing a realistic testbed, including but not limited to, the required distance among strain gauges, the influence of temperature and pipe axial stress. The dataset and analytic results of this work would provide more insights into how to design a realistic testbed for underground PVC water pipe crack detection.

Magnetic induction is a prospective candidate for wireless communication in un-derground water pipelines. It can overcome many challenges of electromagnetic and acoustic methods in the underground environment. However, similar to the other methods, the communication system needs to transfer data efficiently and reliably from underground nodes to above-ground base stations because of ener-gy constraints. Furthermore, difficulties in repairing the underground devices re-quire proper hardware setup to minimise the maintenance. However, the optimal hardware setup and energy efficiency have yet to be studied in a real testbed, leaving a question about the influence of the underground environments on this method. This article presents our work in progress to address the above gaps. In particular, we introduce three questions and corresponding approach methodolo-gy in designing a magnetic induction communication, including the procedure to define an optimal hardware setup, the impact of underground conditions and the tradeoff between energy consumption and communication reliability.

Health care and human comfort are intertwined, but proper health care is costly, especially in the case of dementia. Device-free sensing has the potential to increase safety and comfort in activities of daily living in human lives by enabling truly unobtrusive ways to achieve human activity recognition. However, in order to truly enable device-free sensing in real-life scenarios, it should be integrated seamlessly in wireless communication. Currently, transmitting devices employ high (> 100 Hz) constant transmission rates and the designed neural networks cannot deal with realistic wireless networking, including changing receiver rates.

A premature crack and pipe bending are two of the major signals indicating future failures of a PVC water main. Identifying these defectives when they are young can prevent the break consequence. A new PVC pipe concept that can diagnose its prem-ature crack based on changes in pressure strain characteristics has been introduced. However, the bending influence has yet to be fully explored in this pipe concept. Therefore, this research investigates the full impacts of both bending and a premature crack on the pressure strain characteristic. The result reveals different patterns of pressure strain characteristic deviation between bending and a premature crack. It can help future research to design a method to detect and discriminate these pipe defectives

Speed is an essential parameter in biomechanical analysis and general locomotion research. It is possible to estimate the speed using global positioning systems (GPS) or inertial measurement units (IMUs). However, GPS requires a consistent signal connection to satellites, and errors accumulate during IMU signals integration. In an attempt to overcome these issues, we have investigated the possibility of estimating the horse speed by developing machine learning (ML) models using the signals from seven body-mounted IMUs. Since motion patterns extracted from IMU signals are different between breeds and gaits, we trained the models based on data from 40 Icelandic and Franches-Montagnes horses during walk, trot, t&ouml, lt, pace, and canter. In addition, we studied the estimation accuracy between IMU locations on the body (sacrum, withers, head, and limbs). The models were evaluated per gait and were compared between ML algorithms and IMU location. The model yielded the highest estimation accuracy of speed (RMSE = 0.25 m/s) within equine and most of human speed estimation literature. In conclusion, highly accurate horse speed estimation models, independent of IMU(s) location on-body and gait, were developed using ML.

Traditional modeling approaches, based on predefined business logic, offer little support for today’s complex environments. In this paper, we propose a conceptual agent-based simulation framework to help not only discover complex business processes but also to analyze and learn from emergent behavior arising in cyber-physical systems. Techniques originating from agent-based modeling as well as from the process mining discipline are used to reinforce agent-based decision-making. Whereas agent-technology is used to orchestrate the integration and relationship between the environment and business logic activities, process mining capabilities are mainly used to discover and analyze emergent behavior. Using a functional decomposition approach, we specified three agent types: cyber-physical controller agent, business rule management agent, and emergent behavior detection agent. We use agent-based simulation of a logistics cold chain case study to demonstrate the feasibility of our approach.

The most compelling reason to use unsupervised representation learning as a feature extraction method for effective animal activity recognition is the ability to learn from unlabeled data. Obtaining labeled data is tedious, labor-intensive, and costly, while it is much easier to obtain raw unlabeled data. In this paper, we compare three unsupervised representation learning techniques with three conventional feature extraction methods that are simple and have excellent performance. To investigate the effect of the size of both labeled and unlabeled parts of the dataset on the quality of the representations, we train the representations and classifier using various sample sizes. Furthermore, we evaluate the effect of depth in feature architectures on the performance of the representation learning techniques. All evaluations are performed on two animal datasets that are diverse in terms of species, subjects, sensor-orientations, and sensor-positions. We demonstrate that unsupervised representation learning techniques approach and, in some cases, outperform engineered features in animal activity recognition.

Vibration monitoring and analysis techniques are used increasingly for predictive maintenance. While traditional vibration monitoring relies on wired sensor networks, recent industrial technologies such as WirelessHART, ISA100.11a, and IEEE 802.15.4e have brought a paradigm shift in the automation sector by integrating the flexibility of wireless technologies with the versatility of Internet tools. However, these wireless technologies aren't designed to support strict resource constraint devices, and thus are still unable to fulfill the needs of many modern industrial applications. This article describes a new industrial wireless monitoring system capable of supporting energy-harvested devices through intelligent sensing and network management.

The present invention relates to a method and system for localizing a blind node.

In this paper, we describe and validate the EquiMoves system, which aims to support equine veterinarians in assessing lameness and gait performance in horses. The system works by capturing horse motion from up to eight synchronized wireless inertial measurement units. It can be used in various equine gait modes, and analyzes both upper-body and limb movements. The validation against an optical motion capture system is based on a Bland–Altman analysis that illustrates the agreement between the two systems. The sagittal kinematic results (protraction, retraction, and sagittal range of motion) show limits of agreement of ± 2.3 degrees and an absolute bias of 0.3 degrees in the worst case. The coronal kinematic results (adduction, abduction, and coronal range of motion) show limits of agreement of − 8.8 and 8.1 degrees, and an absolute bias of 0.4 degrees in the worst case. The worse coronal kinematic results are most likely caused by the optical system setup (depth perception difficulty and suboptimal marker placement). The upper-body symmetry results show no significant bias in the agreement between the two systems; in most cases, the agreement is within ±5 mm. On a trial-level basis, the limits of agreement for withers and sacrum are within ±2 mm, meaning that the system can properly quantify motion asymmetry. Overall, the bias for all symmetry-related results is less than 1 mm, which is important for reproducibility and further comparison to other systems.

© ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2018.In most cases, human activity recognition (AR) with smartphones and smartwatches has been done offline due to the limited resources of these devices. Initially, these devices were used for logging sensor data which was later on processed in machine learning tools on a desktop or laptop. However, current versions of these devices are more capable of running an activity recognition system. Therefore, in this paper, we present SmokeSense, an online activity recognition (AR) framework developed for both smartphones and smartwatches on Android platform. This framework can log data from various sensors and can run an AR process in real-time locally on these devices. Any classifier or feature can easily be added on demand. As a case study, we evaluate the recognition performance of smoking with four classifiers, four features, and two sensors on a smartwatch. The activity set includes variants of smoking such as smoking while sitting, standing, walking, biking, as well as other similar activities. Our analysis shows that, similar recognition performance can be achieved in an online recognition as in an offline analysis, even if no training data is available for some smoking postures. We also propose a smoking session detection algorithm to count the number of cigarettes smoked and evaluate its performance.

Shorter hospitalisation is a strategy that healthcare services have adopted to solve the problem of high demand for hospitalisation under conditions of insufficient resources. Early discharge demands patients to self-manage their recovery at home, while recovery is still perceived a clinician’s responsibility. This often results in an extra emotional and psychological load on patients. Due to the lack of measurements to include the emotional and psychological state of a patient in the home context, clinicians often oversee these issues in the evaluation of patients' recovery during hospital visits. This paper outlines the development of esther 1.0, an electronic tool based on Experience Sampling Method, that encourages senior patients of Total Hip Replacement (THR) to self-report their experiences related to their perceived physical and emotional state during the first weeks of recovery at home. This paper describes the emotional assessment of a THR patient and reflects on the value of this knowledge to develop professional support beyond the clinical recovery.

Mobile crowdsensing is power hungry since it requires continuously and simultaneously sensing, processing and uploading fused data from various sensor types including motion sensors and environment sensors. Realizing that being able to pinpoint change points of contexts enables energy-efficient mobile crowdsensing, we modify histogram-based techniques to efficiently detect changes, which has less computational complexity and performs better than the conventional techniques. To evaluate our proposed technique, we conducted experiments on real audio databases comprising 200 sound tracks. We also compare our change detection with multivariate normal distribution and one-class support vector machine. The results show that our proposed technique is more practical for mobile crowdsensing. For example, we show that it is possible to save 80% resource compared to standard continuous sensing while remaining detection sensitivity above 95%. This work enables energy-efficient mobile crowdsensing applications by adapting to contexts.

The main rotor accounts for the largest vibration source for helicopter fuselage and components. However, accurate blade monitoring has been limited due to the practical restrictions on instrumenting rotating blades. The use of Wireless Sensor Networks (WSNs) for real time vibration monitoring promises to deliver a significant contribution to rotor performance monitoring and blade damage identification. This paper discusses the main technological challenges for wireless sensor networks for vibration monitoring on helicopter rotor blades. The first part introduces the context of vibration monitoring on helicopters. Secondly, an overview of the main failure modes for rotor and blades is presented. Based on the requirements for failure modes monitoring, a proposition for a multipurpose sensor network is presented. The network aims to monitor rotor performance, blade integrity and damage monitoring at three different scales referred to as macro layer, meso layer and micro layer. The final part presents the requirements for WSNs design in relation with sensing, processing, communication and actuation. Finally power supply aspects are discussed.

Localization, a process of determining the position of a blind node, can be used in various applications. Signal-strength localization provides a low-cost and lowpower solution to positioning. Signal-strength positioning approaches using fingerprinting or calibrated approaches require a time-consuming calibration phase. Existing self-calibrating approaches, which do not require a priori calibration, use a least-squares fitting model to determine both the position of the blind node as well as the optimal environmental parameters. In this paper, we propose an approach using the Product-Moment correlation between the measured signal strength and the estimated signal strengths. Such approach does not require estimation of the environmental parameters or prior calibration and outperforms existing self-calibrating least-squares approaches. We compare our approach to existing least-squares calibration-free positioning approaches. Moreover, we look at the Cramer-Rao Bound (CRB) of signal-strength localization and using simulations we show that the product-moment correlation outperforms least-squares approaches and follows the CRB closely. Simulation and evaluation using a real-world experiment dataset show the product-moment approach significantly outperforms least-squares approaches. The product-moment approach follows the CRB much more closely and achieves up to twice more accurate positions in certain scenarios. When the error ratio increases and the number of reference positions stays fixed at 6, the product-moment approach scores 20% more accurate positions. In the cooperative localization scenario, the product-moment correlation performs 40% better.

In the context of acoustic monitoring, the location of a sound source can be passively estimated by exploiting time-of-arrival and time-difference-of-arrival measurements. To evaluate the fundamental hardness of a location estimator, the Cramer-Rao bound (CRB) has been used by many researchers. The CRB is computed by inverting the Fisher Information Matrix (FIM), which measures the amount of information carried by given distance measurements. The measurements are commonly expressed as actual distances plus white noise. However, the measurements do include extra noise types caused by time synchronization, acoustic sensing latency, and signal-to-noise ratio. Such noise can significantly affect the performance and depend highly on the sensing platforms such as Android smartphones. In this paper, we first remodel the acoustic-based distance measurements considering such additive errors. Then, we derive a new FIM with the new statistical ranging error models. As a result, we obtain new CRBs for both non-cooperative and cooperative localization schemes that provide better insight into the causality of the uncertainties. Theoretical analysis also proves that the proposed CRBs for localization become the old CRBs when the additional errors are ignored, which gives a robust check for the new CRBs. Thus, the new CRBs can serve as a benchmark for localization estimators with both new and old measurement models. The new CRBs also indicate that there is room to improve current localization schemes, however, it is a daunting challenge.

The proliferation of smartphones nowadays has enabled many crowd assisted applications including audio-based sensing. In such applications, detected sound sources are meaningless without location information. However, it is challenging to localize sound sources accurately in a crowd using only microphones integrated in smartphones without existing infrastructures, such as dedicated microphone sensor systems. The main reason is that a smartphone is a nondeterministic platform that produces large and unpredictable variance in data measurements. Most existing localization methods are deterministic algorithms that are ill suited or cannot be applied to sound source localization using only smartphones. In this paper, we propose a distributed localization scheme using nondeterministic algorithms. We use the multiple possible outcomes of nondeterministic algorithms to weed out the effect of outliers in data measurements and improve the accuracy of sound localization. We then proposed to optimize the cost function using least absolute deviations rather than ordinary least squares to lessen the influence of the outliers. To evaluate our proposal, we conduct a testbed experiment with a set of 16 Android devices and 9 sound sources. The experiment results show that our nondeterministic localization algorithm achieves a root mean square error (RMSE) of 1.19 m, which is close to the Cramer-Rao bound (0.8 m). Meanwhile, the best RMSE of compared deterministic algorithms is 2.64 m.

Applications of body sensor networks (BSN) in health-care are growing rapidly. The number of implanted sensors used to accurately prognose and diagnose the medical conditions are increasing. The implanted medical devices (IMD), for example pace-makers are improvised for delivering patient centric therapies. These IMD require continuous pathological information about the disease from the implanted sensors to ensure the therapeutic value of the treatment. A reliable wireless communication between these heterogeneous medical sensors and devices is essential. The medical sensors and devices have dynamic requirements for wireless communication in terms of throughput, duty-cycle and latency. Fault-tolerant communication links must be devised for medical-emergencies. Wireless communication between the in-body and on-body sensor nodes should be supported. The amount of energy available to the in-body sensor nodes is very limited, which puts forth a strict requirement of ultra-low power consumption. In this work we aim to develop wireless communication protocol which will address the complex requirements of IBSN. This paper briefly explains the research problem and the focus of the research. Preliminary results obtained from the initial work including the characteristics of in-body radio channel and performance of MAC protocols are presented. This paper will identify the research challenges and discuss the planned research approach.

Monitoring the road pavement is a challenging task. Authorities spend time and finances to monitor the state and quality of the road pavement. This paper investigate road surface monitoring with smartphones equipped with GPS and inertial sensors: accelerometer and gyroscope. In this study we describe the conducted experiments with data from the time domain, frequency domain and wavelet transformation, and a method to reduce the effects of speed, slopes and drifts from sensor signals. A new audiovisual data labelling technique is proposed. Our system named RoADS, implements wavelet decomposition analysis for signal processing of inertial sensor signals and Support Vector Machine (SVM) for anomaly detection and classification. Using these methods we are able to build a real time multiclass road anomaly detector. We obtained a consistent accuracy of ≈90% on detecting severe anomalies regardless of vehicle type and road location. Local road authorities and communities can benefit from this system to evaluate the state of their road network pavement in real time.

This paper describes an energy efficiency improvement of the IMA accelerometer-based method for estimating the level of physical activity of a person. The sensor sampling and data processing requirements are significantly reduced by duty-cycling sensor sampling, thus making implementation and long-lasting operation possible on resource-constrained devices as sensor nodes. By duty-cycling, the system maintains adequate bandwidth, while still reducing the effective number of samples taken per unit of time. We analyze in detail the impact of duty-cycling on the accuracy of the method and show that we can reduce the duty-cycle to as little as 10%, incurring a mean error of only about 4%. This translates into energy saving of up to 60% on the sensor node.

We propose a method through which dynamic sensor nodes determine that they move together by communicating and correlating their movement information. We describe two possible solutions, one using inexpensive tilt switches, and another one using low-cost MEMS accelerometers. We implement a fast, incremental correlation algorithm, which can run on resource constrained devices. The tests with the implementation on real sensor nodes show that the method distinguishes between joint and separate movements. In addition, we analyse the scalability from four different perspectives: communication, energy, memory and execution speed. The solution using tilt switches proves to be simpler, cheaper and more energy efficient, while the accelerometer-based solution is more accurate and more robust to sensor alignment problems.

Applications of implantable sensor networks in the health-care industry have increased tremendously over the last decade. There are different types of medium access control (MAC) protocols that are designed for implantable body sensor networks, using different physical layer technologies such as narrow band, ultra wide-band, human body communication, and ultrasound with an innovative low power access technology called wake-up radio (WuR). The WuR operates alongside the main radio either in the same frequency or different frequency, with much lower power and reduced hardware components than main radio. In this article we analyze the impact of WuR on commonly used MAC protocols and evaluate three MAC protocols with WuR using real hardware implanted in animal tissue and compare them with three other MAC protocols without WuR. The hardware implantable board is embedded with a micro-processor, wireless communication unit and is subcutaneously implanted under the skin of the animal tissue. Five nodes with one of them being the central controller connected in star topology are used for evaluation. Energy efficiency, reliability in terms of packet loss ratio, and end-to-end delay for each node are considered as the evaluation criteria.

Context-aware service platforms need to establish and manage trust relationships for users to know if the user's privacy policies are being enforced and for service providers to control access to their services. Current trust solutions are not suitable for this because they do not address in an integrated manner trust issues related to identity provisioning, privacy enforcement and context information trustworthiness. In addition, due to their hierarchical and centralized design, current trust management solutions do not scale well in the ad-hoc pervasive environments in which context-aware platforms are typically deployed. In this paper we propose context-aware trust domains as a management solution for context-aware service platforms.

Localization schemes for wireless sensor networks can be classified as range-based or range-free. They differ in the information used for localization. Range-based methods use range measurements, while range-free techniques only use the content of the messages. None of the existing algorithms evaluate both types of information. Most of the localization schemes do not consider mobility. In this paper, a Sequential Monte Carlo Localization Method is introduced that uses both types of information as well as mobility to obtain accurate position estimations, even when high range measurement errors are present in the network and unpredictable movements of the nodes occur. We test our algorithm in various environmental settings and compare it to other known localization algorithms. The simulations show that our algorithm outperforms these known range-oriented and range-free algorithms for both static and dynamic networks. Localization improvements range from 12% to 49% in a wide range of conditions.

In this paper we identify the common techniques and technologies that are enabling location identification in a ubiquitous computing environment. We also address the important parameters for evaluating such systems. Through this survey, we explore the current trends in commercial products and research in the area of localization. Although localization is an old concept, further research is needed to make it really usable for ubiquitous computing. Therefore, we indicate future research directions and address localization in the framework of our Smart Surroundings project.

Tank water heaters (WHs) are present in a prevailing number of European households. Serving as energy buffers WHs have come under the spotlight of various direct load control (DLC) programs over the last few decades. Although DLC has proven to be an efficient measure towards daily peak demand shaving, the payback effect might lead to a new peak in the grid. This payback phenomenon takes place every time a group of WHs under DLC is permitted to catch up. If not handled properly. This paper presents a novel real-time water flow control approach for domestic water heating systems aiming at decreasing the payback effect of DLC actions. We identify possible control strategies based on an analysis of the water system's thermal dynamics. We formulate the problem of optimal water flow control in terms of minimum WH payback demand and maximum user comfort satisfaction. User comfort is formalized by an integral energy characteristic. Simulations show that water flow control can significantly mitigate the DLC payback effect by reaching the fair compromise between energy savings and discomfort of an end-user.

The present invention relates to a method and a device for localization of a node having an unknown position. More in particular, the present invention relates 5 to a method and a device in which a mobile blind node is used to determine the position of a fixed blind node within a localization area, wherein the fixed blind node is configured to exchange a signal with the mobile blind node.

Ubiquitous location-aware sensing devices have facilitated collection of large volumes of mobility data streams from moving entities such as people and animals, among others. Extraction of various types of periodic behavioral patterns hidden in such large volume of mobility data helps in understanding the dynamics of activities, interactions, and life style of these moving entities. The ever-increasing growth in the volume and dimensionality of such Big Data on the one hand, and the resource constraints of the sensing devices on the other hand, have made not only high pattern recognition accuracy but also low complexity, low resource consumption, and real-timeness important requirements for recognition of patterns from mobility data. In this paper, we propose a method for extracting periodic behavioral patterns from streaming mobility data which fulfills all these requirements. Our experimental results on both synthetic and real data sets confirm superiority of our method compared with existing techniques.

Homecare is believed to alleviate the care responsibilities that professionals and health institutions are facing due to increasing need for care that our current society demands. Relying on the emergence of technological innovations homecare could empower patients to undertake a more active role in the care practice. Understanding the value of technologies to support patients’ awareness and reflection of their own progress is argued as an important step to design appropriate mechanisms that fit patients’ needs. This chapter discusses an integrated patient-centric approach to design homecare technologies considering patients’ personal experiences and social context as crucial aspects when providing care support. This approach proposes the development of in-situ methods and tools to provide a holistic view of patients care experience through the design of interventions in the context of use. The development of ESTHER (Experience Sampling for Total Hip Replacement) is presented to illustrate the benefits and challenges of conducting in-situ research by supporting the collection of physical activity and emotional states in order for patients to self-monitor and self-reflect on the progress of their recovery. The transition from open and simple interventions to more complex and specific ones opens the discussion of the role of technology to introduce new practices that uncover richer experiences by influencing patient’s care activities via self-reporting and self-reflecting mechanisms. The gradual introduction of technologies to uncover from general to more specific experiences it does not only address research needs, more important it brings an understanding on the acceptance of technologies in patients’ care practices. The consequence is a new generation of research to application tools that bring a closer understanding of patients’ experiences and context to inform the design of supportive homecare technologies that embrace a more active role of patients.

Future applications are envisioned to be adaptive to various changes in network, environment and situation. These so-called pervasive applications will be composed from both locally and globally available multimedia resources such as audio and video, web services and context sources. The rapidly increasing pervasiveness in todays networks, i.e. the number of mobile devices, the amount of data they generate and share (near) realtime increases rapidly. In fact this forms the basis for research efforts on the Internet of Things. The increased pervasiveness leads to numerous efficiency and scalability challenges. This chapter will detail the state of the art and binding concepts for efficient realtime sharing and mobility of multimedia and context. Additionally, it lists the associated challenges and their progress in a number of research projects.

Recently the increasing number of sensors integrated in smartphones, especially the iPhone and Android phones, has motivated the development of routing algorithms for Opportunistic Mobile Sensor Networks (OppMSNs). Although there are many existing opportunistic routing algorithms, researchers still have an ambiguous understanding of how these schemes perform on OppMSNs with heterogeneous architecture, which comprises various kinds of devices. In this work, we investigate the performance of well-known routing algorithms in realistic scenarios. To this end, we propose a heterogeneous architecture including fixed infrastructure, mobile infrastructure, and mobile phones. The proposed architecture focuses on how to utilize the available, low cost short-range radios of mobile phones for data gathering and dissemination. We also propose new realistic mobility models and metrics. Selected routing protocols are simulated and evaluated with the proposed heterogeneous architecture, mobility models, and transmission interfaces under various constraints, such as limited buffer size and time-to-live (TTL). Results show that some protocols suffer long TTL, while others suffer short TTL. We further study the benefit of fixed infrastructure in network performance, and learn that most of the opportunistic routing algorithms cannot benefit from the advantage of fixed infrastructure since they are designed for mobile nodes. Finally, we show that heterogeneous architecture need heterogeneous routing algorithms, such as a combination of Epidemic, Spray and Wait, and context-based algorithms.

Vehicular Ad-hoc Networks (VANETs) are expected to serve as support to the development of not only safety applications but also information-rich applications that disseminate relevant data to vehicles. Due to the continuous collection, processing, and dissemination of data, one crucial requirement is the efficient use of the available bandwidth. Firstly, the rate of message transmissions must be properly controlled in order to limit the amount of data inserted into the network. Secondly, messages must be carefully selected to maximize the utility (benefit) gain of vehicles in the neighborhood. We argue that such selection must aim at a fair distribution of data utility, given the possible conflicting data interests among vehicles. In this work, we propose a data dissemination protocol for VANETs that distributes data utility fairly over vehicles while adaptively controlling the network load. The protocol relies only on local knowledge to achieve fairness with concepts of Nash Bargaining from game theory. Simulation results show that our algorithm presents a higher fairness index and yet it maintains a high level of bandwidth utilization efficiency compared to other approaches. In addition, the rate of transmissions is adaptively controlled as new information about the environment is collected.

This paper focuses on optimal and automatic calibration of the propagation model of Received Signal Strength (RSS) based localization algorithms. Conventional RSS-based localization algorithms assume that optimal calibration is static and identical for all nodes, which limits its use to static environments. However realistic environments are dynamic, where each node should estimate its own optimal propagation model settings dependent on the node's hardware and location. We call this process Self-Adaptive Localization (SAL). SAL algorithms estimate the parameter settings from available localization measurements. We show that existing SAL algorithms significantly decrease the localization accuracy and stability. Our main contribution is that we determine the conditions under which SAL algorithms provide optimal results, that are shown to be constraints on the localization surface. Since the antenna orientation has a significant impact on RSS and thus optimal propagation model settings, we evaluated SAL in an environment with unknown and thus dynamic antenna orientations. Our measurements and simulations show that these constraints increase the accuracy by ~45% and the stability by ~70% in static and dynamic environments.

Vehicular Sensor Networks (VSNs) are an emerging area of research that combines technologies developed in the domains of Intelligent Transport Systems (ITS) and Wireless Sensor Networks. Data dissemination is an important aspect of these networks. It enables vehicles to share relevant sensor data about accidents, traffic load, or pollution. Several protocols are proposed for Vehicle to Vehicle (V2V) communication, but they are prone to intermittent connectivity. In this chapter, the authors propose a roadside infrastructure to ensure stable connectivity by adding vehicle to infrastructure to the V2V communication. They introduce a data dissemination protocol, Hexagonal Cell-Based Data Dissemination, adapting it for VSNs within a metropolitan area. The virtual architecture of the proposed data dissemination protocol exploits the typical radial configuration of main roads in a city, and uses them as the basis for the communication infrastructure where data and queries are stored. The design of the communication infrastructure in accordance with the road infrastructure distributes the network data in locations that are close or easily reachable by most of the vehicles. The protocol performs a geographical routing and is suitable for highly dynamic networks, supporting a high number of mobile sources and destinations of data. It ensures reliable data delivery and fast response. The authors evaluate the performance of the proposed protocol in terms of data delivery ratio and data delivery delay. The simulation results show that HexDD significantly improves the data packet delivery ratio in VANETs. Publisher's Version

Time-critical applications of Wireless Sensor Networks (WSNs) demand timely data delivery for fast identification of out-of-ordinary situations and fast and reliable delivery of notification and warning messages. Due to the low reliable links in WSNs, achieving real-time guarantees and providing reliable data is quite challenging. To ensure data reliability, traditionally various retransmission mechanisms have been used, which in turn introduce extra delay. In this paper, we propose READ, i.e., a reliable and realtime aggregation-aware data dissemination to ensure reliable and fast data delivery in a chain-based WSN. We also investigate the relatively unexplored topic of impact analysis of Time To Live (TTL) and link reliability parameters on network performance in terms of attained hit ratio for three different approaches, i.e., READ, QoS-ACA, and the stop-andwait (S-W) ARQ to assess the appropriateness of each method facing different conditions. The simulation results show READ performs better in terms of hit ratio compared with QoS-ACA and S-W ARQ when link reliability is low and packet's TTL is short. Although not being the primary goal of READ, energy consumption of the protocol is also much lower than the other two approaches.

Recently, there has been a growing interest within the research community in developing opportunistic routing protocols. Many schemes have been proposed; however, they differ greatly in assumptions and in type of network for which they are evaluated. As a result, researchers have an ambiguous understanding of how these schemes compare against each other in their specific applications. To investigate the performance of existing opportunistic routing algorithms in realistic scenarios, we propose a heterogeneous architecture including fixed infrastructure, mobile infrastructure, and mobile nodes. The proposed architecture focuses on how to utilize the available, low cost short-range radios of mobile phones for data gathering and dissemination. We also propose a new realistic mobility model and metrics. Existing opportunistic routing protocols are simulated and evaluated with the proposed heterogeneous architecture, mobility models, and transmission interfaces. Results show that some protocols suffer long time-to-live (TTL), while others suffer short TTL. We show that heterogeneous sensor network architectures need heterogeneous routing algorithms, such as a combination of Epidemic and Spray and Wait.

Physical activity, therapy and exercising are essential for the management of many chronic conditions. This is why human motion tracking for healthcare applications has received significant attention recently. In this chapter we describe the architecture in which we provide a person-centric healthcare solution for patients with chronic conditions based on the recent advances in wireless inertial sensing systems. We emphasize the role of the home as care environment, by providing real-time support to patients in order to monitor, self-manage and improve their physical condition according to their specific situation.

Online processing is essential for many sensor network applications. Sensor nodes can sample far more data than what can practically be transmitted using state of the art sensor network radios. Online processing, however, is complicated due to limited resources of individual nodes. The naive Bayes classifier is an algorithm proven to be suitable for online classification on Wireless Sensor Networks. In this paper, we investigate a new technique to improve the naive Bayes classifier while maintaining sensor network compatibility. We propose the application of unsupervised learning techniques to enhance the probability density estimation needed for naive Bayes, thereby achieving the benefits of binning histogram probability density estimation without the related memory requirements. Using an offline experimental dataset, we demonstrate the possibility of matching the performance of the binning histogram approach within the constraints provided by Wireless Sensor Network hardware. We validate the feasibility of our approach using an implementation based on Arduino Nano hardware combined with NRF24L01+ radios.

This paper presents a novel Radio Interferometric Positioning System (RIPS), which we call Stochastic RIPS (SRIPS). Although RIPS provides centimeter accuracy, it is still not widely adopted due to (1) the limited set of suitable radio platforms and (2) the relatively long measurement and calibration times. SRIPS overcomes these practical limitations by (1) omitting the calibration phase of the existing RIPS and by (2) applying a novel positioning algorithm. SRIPS exploits the phenomenon of the small but stable difference between two transmitted frequencies that often exists when two radios are tuned to the same frequency. We obtain an experimental measure for this stability. This approach enables the implementation of RIPS on commonly available radio platforms, such as the CC2430, because fine-tuning in small steps relative to the beat frequencies for calibration is not required. In addition, we show that SRIPS calculates the position that provides the best fit to the set of measurements, given the underlying statistical and propagation models. Therefore, SRIPS converges more accurately to the true locations in a variety of situations of practical interest. Experiments in a 20x20m2 set-up verify this and show that our SRIPS CC2430 implementation reduces the number of required measurements by a factor of three, and it reduces the measurement time to less than 0.1 seconds, while providing accuracy similar to that of the existing RIPS implementation on the CC1000 platform, which requires seconds.