Ubiquitous Computing and Ambient Intelligence

The suitable operation of mobile robots when providing Ambient Assisted Living (AAL) services calls for robust object recognition capabilities. Probabilistic Graphical Models (PGMs) have become the de-facto choice in recognition systems aiming to efficiently exploit contextual relations among objects, also dealing with the uncertainty inherent to the robot workspace. However, these models can perform in an incoherent way when operating in a long-term fashion out of the laboratory, e.g. while recognizing objects in peculiar configurations or belonging to new types. In this work we propose a recognition system that resorts to PGMs and common-sense knowledge, represented in the form of an ontology, to detect those inconsistencies and learn from them. The utilization of the ontology carries additional advantages, e.g. the possibility to verbalize the robot’s knowledge. A primary demonstration of the system capabilities has been carried out with very promising results.

This paper addresses a practical and challenging problem concerning the recognition of behavioral symptoms dementia (BSD) such as aggressive and agitated behaviors. We propose two new algorithms for the recognition of these behaviors using two different sensors such as a Microsoft Kinect and an Accelerometer sensor. The first algorithm extracts skeleton based features from 3D joint positions data collected by a Kinect sensor, while the second algorithm extracts features from acceleration data collected by a Shimmer accelerometer sensor. Classification is then performed in both algorithms using ensemble learning classifier. We compared the performance of both algorithms in terms of recognition accuracy and processing time. The results obtained, through extensive experiments on a real dataset, showed better performance of the Accelerometer-based algorithm over the Kinect-based algorithm in terms of processing time, and less performance in terms of recognition accuracy. The results also showed how our algorithms outperformed several state of the art methods.

This paper introduces the description of the design and operation of a motorized multi-camera slider. The slider has been developed as part of an assistive technology aimed at monitoring physical rehabilitation exercises from different views. The monitoring of the exercises is performed from a couple of RGB-D Kinect cameras that are placed on the most convenient positions of the slider in an intelligent manner to capture the best views of the patient’s important body parts. This paper focuses mainly on the description of the dynamic model of the platform and the control of the motorized slider.

Medicine research based in machine learning methods allows the improvement of diagnosis in complex diseases. Age related Macular Degeneration (AMD) is one of them. AMD is the leading cause of blindness in the world. It causes the 8.7 % of blind people. A set of case and controls study could be developed by machine-learning methods to find the relation between Single Nucleotide Polymorphisms (SNPs) SNP_A, SNP_B, SNP_C and AMD. In this paper we present a machine-learning based analysis to determine the relation of three single nucleotide SNPs and the AMD disease. The SNPs SNP_B, SNP_C remained in the top four relevant features with ophthalmologic surgeries and bilateral cataract. We aim also to determine the best set of features for the classification process.

Ambient displays provide peripheral awareness of an event or situation of importance, but not critical. One such situation is monitoring the sleep of people with dementia (PwD). They frequently exhibit problematic behaviors at night, such as waking up often or wandering, which cause significant burden to caregivers. In this paper we propose the use of ambient displays for caregivers in a geriatric residence to be aware of the sleep of PwD. We report the results of a user study conducted in a geriatric residence to determine potential applications of ambient displays. From an ethnographic analysis of the data gathered in the study we derive design insights and application scenarios. We validated these results with caregivers who work in two geriatric residences.

The way to achieve enough data for data mining is by accessing existing databases or directly acquiring data with the aim of creating new databases. In this paper we present the DAFIESKU system built to acquire different types of physiological data via experiments and the factors taken into account when developing it, in order to facilitate the creation of new datasets by means of mobile and wearable devices. DAFIESKU has been evaluated on a case study associated to non-classroom learning.

The significant increase of ageing population calls for solutions that help the elderly to live an independent, healthy and low risk life, but also ensure their social interaction. The improvements in Information and Communication Technologies (ICT) and Ambient Assisted Living (AAL) have resulted in the development of equipment that supports ubiquitous computing, ubiquitous communication and intelligent user interfaces. The smart home technologies, assisted robotics, sensors for health monitoring and e-health solutions are some examples in this category. Despite such growth in these individualized technologies, there are only few solutions that provide integrated AAL frameworks that interconnect all of these technologies. In this paper, we discuss the necessity to opt for an integrated solution in AAL. To support the study we describe real life scenarios that help us justify the need for integrated solutions over individualized ones. Our analysis points to the clear conclusion that an integrated solution for AAL outperforms the individualized ones.

In our current work we propose a strategy to reduce the vast amounts of data produced within smart environments for sensor-based activity recognition through usage of the nearest neighbor (NN) approach. This approach has a number of disadvantages when deployed in resource constrained environments due to its high storage requirements and computational complexity. These requirements are closely related to the size of the data used as input to NN. A wide range of prototype generation (PG) algorithms, which are designed for use with the NN approach, have been proposed in the literature to reduce the size of the data set. In this work, we investigate the use of PG algorithms and their effect on binary sensor-based activity recognition when using a NN approach. To identify the most suitable PG algorithm four datasets were used consisting of binary sensor data and their associated class activities. The results obtained demonstrated the potential of three PG algorithms for sensor-based activity recognition that reduced the computational complexity by up to 95 % with an overall accuracy higher than 90 %.

Dynamic Instance Activation (DIA) is a newly developed data-driven classification algorithm. It was designed to minimise the negative impact in situations of data incompleteness and inconsistency. To achieve this, the proposed methodology attempts to maximise the accuracy of the classification process in a way that does not compromise the overall computational effort. In this research, DIA was evaluated in the context of Human Activity Recognition (HAR) for Smart Environments, using datasets consisting of binary sensor data and their associated class labels (activities). This scenario was selected as an ideal case study to illustrate the usefulness of DIA considering the wide range of domains in which HAR is applied. It was also considered adequate given the simplicity of the data involved in the process, which allows using relatively simple similarity functions, therefore placing the main focus on DIA’s performance.In this context, DIA was compared with other state-of-the-art classifiers, delivering promising results in terms of percentage of activities correctly identified over the total. It is important to note that these results could be further improved if other similarity functions or data representation schemes were selected.

Accidental falls can cause serious injury to at risk individuals. This is especially true in the elderly community where falls are the leading cause of hospitalization, injury-related deaths and loss of independence. Detecting and rapidly responding to falls has shown to reduce the long-term impact of and risks associated with falls. A number of real time fall detection solutions exist, however, these have some deficiencies relating to privacy, maintenance, and correct usage. This study introduces a novel fall detection approach that aims to address some of these deficiencies through use of computer vision processes and ceiling mounted thermal vision sensors. A preliminary evaluation has been performed on this process showing promising results, with an accuracy of 68 %, however, highlighting a number of issues related to false positives. Future work will improve this approach and provide extended evaluation.

The development of novel activity recognition approaches requires access to high quality, annotated datasets. When designing an experiment, researchers may not have access to the complete set of equivalent sensors required. A viable solution to this barrier has been the use of completely simulated environments. Nevertheless, an optimal solution may be to allow researchers to use the equipment they do have, and simulate missing sensors. This paper aims to address this scenario through the proposal of the Intelligent Environment Experiment Assistance Tool. The approach facilitates real-time partial simulation of an environment in addition to a real-time annotation component which aims to maximize the accuracy and consistency of dataset annotations. The concept has received feedback from 19 international researchers who are involved in Intelligent Environments research. This feedback provides an insight into the prevalence, type and impact of limitations in physical IE usage. Additionally, 84.3 % of the researchers stated that the real-time annotation would be very useful or quite useful, and 52.6 % stated that the partial simulation component would be very useful or quite useful.

In this paper we study the use of medical history information extracted from the Utah Population Database (UPDB) to predict adoption of a reminder solution for people with dementia. The adoption model was built using 24 categorised features. The kNN classification algorithm gave the best performance with 85.8 % accuracy. Whilst data from the UPDB is more readily available than that in our previous work, the results highlight the benefit of including psychosocial and background information within an adoption model.

It is fully appreciated that progress in the development of data driven approaches to activity recognition are being hampered due to the lack of large scale, high quality, annotated data sets. In an effort to address this the Open Data Initiative (ODI) was conceived as a potential solution for the creation of shared resources for the collection and sharing of open data sets. As part of this process, an analysis was undertaken of datasets collected using a smart environment simulation tool. A noticeable difference was found in the first 1–2 cycles of users generating data. Further analysis demonstrated the effects that this had on the development of activity recognition models with a decrease of performance for both support vector machine and decision tree based classifiers. The outcome of the study has led to the production of a strategy to ensure an initial training phase is considered prior to full scale collection of the data.

Sensor technology for personalized physiologic monitoring contributes to health and safety in workplaces. Wearable devices represent an efficient way to capture physiological values to obtain individual efforts for each worker due to physical activities. Heart rate based ergonomic methods provide results that show drudgery of a work activity for each person. In this paper, we show some experiments that highlight the importance when using custom methods to estimate the effort of people doing physical work. Previous works have already validated some benefits in using sensor technology to estimate physical efforts and energy consumption in the workplace. The results in our experiments applied to cleaning staff demonstrate how important is to use personalized measurements for more objective effort estimation.

Due to their flexibility and immediacy, participatory sensing systems based on ad hoc networks have become a valuable source of proximity-based information about events and incidents in the city. However, easy contribution may result in the transmission of larges amount of data, some of them being duplicate or not relevant. We have proved that collaboratively filtering the information to be disseminated improves network efficiency while maintaining the system effectiveness.

In this paper, a localization system of mobile nodes in a ZigBee Home Automation (ZHA) network has been developed. We used the ZigBee wireless protocol for networking due to its low cost, low power consumption, and acceptable data rate for most of smart home control systems, compared to other existing alternatives like Wi-Fi, Bluetooth or ZWave. Between the numerous localization techniques, we used the Received Signal Strength Indicator (RSSI), which can be obtained without the inclusion of additional hardware to the ZigBee nodes. Our system was implemented with three different Artificial Neural Networks (ANN), which results are compared and analysed; with a series of data treatments to avoid communications errors between the nodes and RSSI values fluctuations.

Sensor networks are becoming popular nowadays in the development of smart environments. Heavily relying on static sensor and actuators, though, such environments usually lacks of versatility regarding the provided services and interaction capabilities. Here we present a framework for smart environments where a service robot is included within the sensor network acting as a mobile sensor and/or actuator. Our framework integrates off-the-shelf technologies to ensure its adaptability to a variety of sensor technologies and robotic software. Two pilot cases are presented as evaluation of our proposal.

This work addresses content-based publish/subscribe in dynamic ubiquitous scenarios, composed of a variety of heterogeneous devices connected by wireless networks, mobility being an intrinsic characteristic of this kind of systems. We present an approach based on the simple routing strategy, which is incrementally extended in order to support client and broker mobility. We allow both client and broker migration without the need to flood the network to notify of this change.

Underwater sensor networks are an important field of research. Several applications require the use of this kind of networks like tsunami or oil spill alerts. The underwater medium is very harsh and only acoustic signals can be used for the transmission of information. The use of this kind of networks is still in a developing state far from reaching standard consensus on basic aspects like carrier frequency or modulation techniques. The use of these networks for real-time applications has not been analyzed previously. In this paper we present two solutions for the scheduling of real-time messages and we provide a time constraint analysis of the performance of the network.

This work describes a novel solution to face a civil security problem in the area of search and rescue of missing people. This proposal is based on the use of emerging technologies like unmanned aerial vehicles, also known as drones, and the use of simulated beacons in smartphones. In particular, the solution presented here which is focused on the use of several synchronized drones going over an area and scanning it to check whether they detect any Bluetooth Low Energy beacon signal emitted from the smartphone of a missing person and retrieve the exact GPS position. This paper also includes a preliminary study of possible attacks on the different elements and the security measures used to protect them.

The rise in low power devices has created a necessity for connectivity among systems, in some cases across great geographic lengths. Current wireless communication protocols for these devices cover only small areas or require several hops to communicate. LPWAN have surfaced to cover this necessity offering greater transmission range in energy efficient protocols. Among these can we find the LoRa technology, offering wide spread spectrum modulation for WSN. The most common implementations of LoRa work at 868 MHz frequency range, and there is few information for other frequency bands. Given the nature of this modulation, it is of great interest to analyse lower frequencies. In this article a direct comparison of LoRa in the 868 MHz and 433 MHz will be done. Additionally, several parameters will be modified to find the best configuration available. This will help validate the possibility of transmitting at longer distances than current 868 MHz implementations.

The interaction between electromagnetic waves and matter has led to the development of applications to detect and characterise them. The conventional systems use the emission, transmission and reception of waves at a specific frequency range to detect medium parameters (constant dielectric, permittivity, conductivity or permeability) of an analysed area. The interaction between the electromagnetic wave and the analysed medium depends on the range of frequency used. This phenomenon is used in different disciplines and working environments, geoscience or medical disciplines are examples where the use of electromagnetic waves provides non-intrusive applications with clear benefits. Each frequency of signal transmitted and received is analysed to determine the interaction produced in absolute measurements. In this work a method based in differential measurement technique is proposed as a novel way of detecting and characterizing electromagnetic matter characteristics. The theoretical results show that it is possible to obtain benefits from the behaviour of the wave-medium interaction using differential measurement on reception of electromagnetic waves at different frequencies. Differential measures introduce advantages in detection processes and increase development possibilities of new non-intrusive applications.

This work presents an in-depth analysis of the influence of different configuration parameters of the wireless sensor network routing protocol, DARP. These parameters regulate the protocol performance in different scenarios. The analysis is based on an ANalysis Of VAriance (ANOVA) study of the effect on the performance of the configuration parameters of DARP. The objective of this work is to identify the most relevant configuration parameters, therefore finding the most effective values for a Smart City-type scenario.

This paper proposes a self-adaptive sampling scheme for WSNs, which aims at capturing accurately the behavior of the physical parameters of interest in each specific WSN context yet reducing the overhead in terms of sensing events. The sampling scheme relies on a set of low-complexity rules capable of auto-regulate the sensing frequency in accordance with each parameter behavior. As proof-of-concept, based on real environmental datasets, we provide statistical indicators illustrating the added value of the proposed sampling scheme in reducing sensing events without compromising the estimation accuracy of physical phenomena.

New technologies, such as machine-to-machine networking, and a demand for energy-efficient buildings and security are driving the building automation market forward. The use of low cost wireless sensor networks (WSN) provides many advantages for sustainable building automation solutions. In this paper, we analyzed the WSN related requirements for sustainable building automation operations and maintenance. We used the three sustainability dimensions, economic, environment and social, to gather the requirements. The requirement analysis resulted in 10 requirements aiming at maximization of energy conservation, lowering of operation and maintenance costs while taking into accounts the users’ or occupants’ preferences, comfort, and safety.

In this paper we propose a routing algorithm for Mobile Wireless Sensor Network (MWSN) where all the nodes, sensors and sink, are mobile. We follow a leader election approach to build a spanning tree by using only information provided by neighbor nodes. The leader election approach allows automatic routing adaptation despite node mobility. The routing algorithm has been implemented and evaluated using the JBotSim simulator. We present some preliminary evaluations that show the impact of mobility in the message delivery rate.

Field operations carried out by mobile users in inhospitable contexts like disaster relief or battlefield require that the communication between those mobile users is as effective and efficient as possible. To facilitate such communication, a highly available communication platform is required. This article envisions that Mobile Ad-hoc Networks (MANETs) could be feasible communication platforms for inhospitable contexts provided high connectivity is achievable. In this article, a proactive distributed mechanism, referred to as a self-organizing mechanism, to provide high connectivity for MANETs operating in inhospitable contexts is presented. This mechanism is based on the introduction of additional unmanned vehicles acting as communication bridges, where their demand is dynamically determined by the movement of the mobile users. By means of simulations it is demonstrated that the proactive distributed self-organizing mechanism proposed in this article allows MANETs to exhibit high connectivity under high values of link failure probability.

Driver distraction refers to the lack of attention to the driving tasks due to engagement in secondary tasks. Most methods reported in the literature are based on visual-features analysis of head pose, since it is a strong indication of driver distraction. In contrast, we propose to use the inertial sensors embedded in smart glasses. To this aim, we collected data from five participants and conducted experiments to assess the feasibility of using support vector machines (SVM) to generate drivers’ models to infer their focus of attention. The results show that using the personalized training model renders an acceptable accuracy to identify particular car cabin’s spots where drivers focus their attention (i.e. accuracy was greater than 50 % and less than 81.44 % for all subjects).

Seamless integration of devices in Ubiquitous Computing requires partaking nodes to perform periodic scans for new nodes in the area, and a set of protocols for interchanging information between devices, among them Bluetooth. In this paper we show an evaluation of the Bluetooth Serial Port Profile (SPP) to study how the communication performance behaves under the discovery process of Bluetooth for new nodes in the vicinity. The results so far show that the performance loss is not dramatic but that it can be improved using different strategies.

Industry 4.0 or cyber-industry may employ Cyber-Physical Systems (CPS) profusely to organize production media in a new and more efficient way. Many processes might be defined in CPS and Industry 4.0, however, physical processes are probably the least studied. Therefore, in this paper we propose an architecture for Industry 4.0-based systems focused on the control of physical processes. The architecture follows the CPS paradigm. Moreover, a first evaluation of the performance of the proposed solution, using simulation tools, is provided. The results proved the proposed architecture is a valid solution for physical control processes.

In this paper we describe the advantages of using Near Field Communication technology (NFC) for engagement strategies carried out by companies or institutions. Augmented objects with NFC chips (gadgets) are given to potential targets having associated appropriate applications without requiring the installation of mobile applications. These gadgets and applications serve as bait for companies and institutions to be able to communicate with targets and spread any kind of multimedia aims for their engagement. We also describe the Red Thread application, a solution based on a romantic legend which is very suitable for young people whom universities want to attract every year for new enrolments.

A typical application scenario related to the Internet of Things consists of smart sensing devices collecting and transmitting measurement data from their surroundings to a back-end. Efficiently deploying such devices on the field for the purposes of such applications is not a trivial task. Deployment of current IoT and sensor solutions on the field are often time consuming and inconvenient. Setup and configuration of both wireless and wired solutions require know-how, training and time from the person carrying out the deployment of the devices. In order for the rising amount of IoT-based solutions to break into wider usage, a solution for quick and easy deployment is needed that is both easy-to-understand and configuration free. In this work, a proposed solution for rapid deployment of IoT devices is prototyped and presented.

The Internet of Things (IoT) imposes several challenges and opportunities. Most of the existing and ongoing network solutions are limited in scope and technology, ignoring important IoT aspects and using traditional protocols and technologies that are not necessarily suitable for things. The proposal is named the Advanced Network of Things (ANT) which constitutes an alternate solution to provide network enhanced services in IoT. In this paper the ANT model design is presented.

The number of Internet of Things systems in which the general population is involved keeps increasing. At the same time, the configuration and adaptation of these systems to their users’ preferences are still mostly manual. To achieve a better personalization of these systems comprehensive virtual profiles of their users, including as much information as possible, is needed. One of the most relevant information of a user virtual profile is his location. With the rise of the bluetooth low energy protocol and the appearance of the beacon technology, indoor positioning of smartphone owners can be added to these virtual profiles. However, the lack of a beacon standard and the low level information provided by these devices is hindering its adoption. To solve this, this paper presents a Beacon Management System that abstracts applications developers from the low level details of the different beacons protocols and allows the management of unregistered beacons. The proposed system has been integrated into the People as a Service computation model and in the commercial platform nimBees to improve the creation of virtual profiles in IoT applications.

The use of automated surveillance systems has become essential in different application fields, from the domestic to the military. Current surveillance systems used in controlled environments can be easily complemented through the use of robots to improve their functionality. This paper describes RoboCAM, a secure video surveillance application for Android devices based on robots. The proposal uses a smartphone placed on a Lego Mindstorms EV3 robot and connected via Bluetooth to it. The smartphone runs in background a motion detection application with its camera. If motion is detected, it begins to capture images that are sent to a streaming server, in order to see them live from a web application. Remotely via the web application and the smartphone connected via Bluetooth it is possible to move the robot in order to find out more information about what is happening in the monitored location.

This work presents a comparative study between two of the most used streaming protocols, RTSP and WebRTC. It describes a system designed to evaluate times at live streaming: establishment time and stream reception time from a single source to a large quantity of receivers with the use of smartphones. Two systems that use the commented protocols have been implemented, specifically, two Android applications that use these protocols in the scope of video surveillance at airports. Both systems are composed of a mobile application and a web service. The design of the systems has been done avoiding differences between these protocols for P2P communication in the same local network. Several simulations have been performed to compare them and the obtained results have been used for a comparative study between streaming establishment and sending package times of each protocol.

To achieve energy self-sufficiency in housing, it is necessary to know the capacity of microgeneration of systems and the real needs of consumption of the users. Each house includes a microgeneration system based on the use of sustainable energy. In theoretical conditions, the design is capable of satisfying entirely its energy needs. However in real conditions there are situations in which it is necessary more energy, and others on the contrary, it produces more energy than necessary. This paper proposes the use of context-aware paradigm that integrate ubiquitous sensor network for the optimization of the energy resource available. The goal is to achieve the maximum energy efficiency during the distribution, in real time and according to energy demand in housings and community equipments. Using context-aware paradigm the system obtains environmental information, determines the profile of consumption of the users and calculates the instantaneous energy needs. This information is obtained from ubiquitous networks and will be used by systems of distribution of the electrical grid of the group of housings. Different paradigms of communication are used that provides the necessary support to achieve near real time. These paradigms are based on the protocols used in the development of the Internet of things.

Data communication is at the heart of any distributed system. The adoption of generic data formats such as XML or JSON eases the exchange of information and interoperability among heterogeneous systems. However, the verbosity of those generic data formats usually requires system resources that might not be available in resource-constrained systems, e.g., embedded systems and those devices which are being integrated into the so-called IoT. In this work we present a method to reduce the cost of managing data models like XML or JSON by using templates and context based compression. We also provide a brief evaluation and comparison as a benchmark with current implementations of W3C’s Efficient XML Interchange (EXI) processor. Although the method described in this paper is still at its initial stage, it outperforms the EXI implementations in terms of memory usage and speed, while keeping similar compression rates. As a consequence, we believe that our approach fits better for constrained systems.

The Internet of Things (IoT) is an emergent paradigm characterized by a plethora of smart objects connected to the Internet. An inherent characteristic of IoT is the high heterogeneity and the wide distribution of objects, thereby calling for ways to describe in an unambiguous and machine-interpretable way the resources provided by objects, their properties, and the services they offer. In this context, discovery services play a significant role as they allow clients (middleware platforms, end-users, applications) to retrieve available resources based on appropriate search criteria, such as resource type, capabilities, location, and Quality of Context (QoC) parameters. To cope with these concerns, we introduce QoDisco, a QoC-aware discovery service relying on multiple-attribute searches, range queries, and synchronous/asynchronous operations. QoDisco also comprises an ontology-based information model for semantically describing resources, services, and QoC-related information. In this paper, we describe the QoDisco architecture and information model as well as an evaluation of the search procedure in an urban air pollution monitoring scenario.

A lot of devices for ubiquitous computing are portable and can be used in any location. Therefore energy storage systems become increasingly important to power mobile devices. Supercaps provide a high power density, hence they are, for example, suited for driving high current peaks that occur during wireless transmissions. However, an analysis of first generation devices revealed that permanent application of the nominal voltage stressed the supercap to such an extent that the life cycle was unexpectedly short. Now a further analysis of second generation supercaps regarding reliability and life cycle will try to find out if the overall performance has improved.

The design of an architecture of communication oriented to medical and sports applications is presented, with the goal to get a series of information collected from a network of portable sensors, those sensors are arranged in the body of the person in order to gather vital constants of the user and then transmit them to a web service on the internet, as well as to monitor the state of health and user’s physical performance.

There is an increasing need for personalised and context-aware services in our everyday lives and we rely on mobile and wearable devices to provide such services. Context-aware applications often make use of machine-learning algorithms, but many of these are too complex or resource-consuming for implementation on some devices that are common in pervasive and mobile computing. The algorithm presented in this paper, named CAMP, has been developed to obtain a classifier that is suitable for resource-constrained devices such as FPGA:s, ASIC:s or microcontrollers. The algorithm uses a combination of the McCulloch-Pitts neuron model and Cellular Automata in order to produce a computationally inexpensive classifier with a small memory footprint. The algorithm consists of a sparse binary neural network where neurons are updated using a Cellular Automata rule as the activation function. Output of the classifier is depending on the selected rule and the interconnections between the neurons. Since solving the input-output mapping mathematically can not be performed using traditional optimization algorithms, the classifier is trained using a genetic algorithm. The results of the study show that CAMP, despite its minimalistic structure, has a comparable accuracy to that of more advanced algorithms for the datasets tested containing few classes, while performing poorly on the datasets with a higher amount of classes. CAMP could thus be a viable choice for solving classification problems in environments with extreme demands on low resource consumption.

Water management systems are having an increasing impact for sustainable development on all types of urban infrastructures. This paper describes the framework of a water management system, as a project of transformation from a traditional commercial software approach to a new, open source flexible system, capable to afford new demanding requirements. Among these new demanding requirements we cover analysis and implementation issues for: an open source GIS for water management, the challenge to big data management, reusing existing data, and network integrity and robustness. A real case scenario is thoroughly described. The integration of a powerful data model and spatial process model into a water modeling framework is proved to enhance the software capabilities for water management system.

Nowadays the public transport systems play a main role in the advanced societies. How to evaluate the quality of the public transport is a critic task of the transport regulatory agencies. One indicator used to measure this quality is the fulfilment of the scheduled operation by the transport operators, specially scheduled arrival times and frequencies at stops. In this paper, an automatic system to estimate arrival times in the context of road public transport is proposed. The system works autonomously, acquiring massive vehicle position readings, registering and processing them automatically in order to estimate arrival times. This autonomous behaviour is achieved using pattern recognition and statistical techniques. To illustrate the application of the system, the estimation of arrival times at a bus stop is presented.

Given their ubiquity and sensing capabilities, current smartphones have been used to explore different real-life tracking and monitoring scenarios. Particularly, in the domain of Intelligent Transportation Systems (ITS), the exact orientation of the smartphone must be known to gain full advantage of the data provided by its internal accelerometers. From here, rich contextual information could be inferred. Nonetheless, in real-life scenarios, smartphones are freely placed within vehicles, so a reorientation strategy needs to be applied. The usage of several algorithms to reorient acceleration readings has been mentioned for ITS applications, but very little evaluation of their efficacy has been performed. In this work, we study the effectiveness of four algorithms for vertical reorientation, and two for triaxial reorientation of acceleration readings. Results suggest that all methods for the vertical case are equivalent, however, in the case where triaxial reorientation is needed, current strategies are far from acceptable results. We expect these findings could promote further research to alleviate these issues.

The objective of this work is to synthesize the difficulties an algorithm must handle in book digitization for subsequent OCR application, such as angle correction, image distortion and words segmentation in addition to being operated by blind or visually impaired people real-time by video stream without further assistance. The developed method seems reliable, and provides good OCR results on a page by page basis. The results show improvements above 99,3 % in OCR performance in some cases, although execution time has increased. “The Vocalizer Project” emerged from a demand from the Brazilian Ministry of Culture and Education for application in schools and public libraries. It aims to create more inclusive smart cities. Furthermore, it is destined for the inclusion of visually impaired and blind people to the vast bibliographic material existent.

The literature has offered a number of surveys regarding the concept of smart city, but few assessments of toolkits. This paper presents a short analysis of existing smart city toolkits. The analysis yields some general observations about existing toolkits. The article closes with a brief introduction of the Open City Toolkit, a toolkit currently under development which aims at addressing some of the gaps of existing toolkits.

The natural and built environment has an impact on the amount of daily physical activity that people do, being less than the minimum recommended by international organizations. However, some characteristics of the environment may incentive physical activity more than others. While technological advances in geospatial information can be regarded as powerful enablers to identify and analyse these main features of the built environment, there is still a large gap between health-related professionals and the proper use of such technologies to address the prevalence of physical inactivity. This paper presents the research plan being conducted for a doctoral thesis with two major objectives: to identify a Playability Index that gives insights about how friendly is a city from the perspective of physical activity, and to develop a technological framework that eases the development of software-driven, effective interventions for promoting physical activity for young people in urban areas.

Mobility is a basic need for people, therefore the authorities in advanced societies pay special attention to systems and initiatives that improve or facilitate it. In this work, a ubiquitous signaling system for visually impaired people, using a combination of WiFi and Bluetooth Low Energy (BLE) technology with Smart Mobile Devices, is proposed. The system is conceived as a solution to help and support these users in places of public road transportation with travellers massive transit, such as bus stations, interchangers or preferred stops. In addition, once the user is positioned in the desired location, the system would offer advanced services: information on lines schedules, tickets sale, etc.

Lowfield Park in Sheffield, UK is a green recreational space maintained by the City Council. Lowfield Park was selected as the primary Sheffield FieldLab for the ProFit project which ended in 2015. The ProFit project was European Interreg IVbNWE funded with the aim of encouraging physical activity through innovations in products, services and ICT systems. In 2014 the Sheffield Hallam University City Athletics Stadium (SHUCAS) was introduced as a secondary FieldLab. A number of innovative systems have been installed into the FieldLabs, these include: Pan Tilt Zoom cameras, automatically timed sprint and running tracks, outdoor displays/touchscreen and a gait analyser. This paper describes the hardware, software and cloud infrastructure created to enable these systems. Pilot testing has been carried out over the last year and has found a positive effect on both sites. The systems created will be taken forward to Sheffield’s Olympic Legacy Park, which is currently under development.

This paper describes the WeLive framework, a set of tools to enable co-created urban apps by means of bringing together Open Innovation, Open Data and Open Services paradigms. It proposes a more holistic involvement of stakeholders across service ideation, creation and exploitation. The WeLive co-creation process applied to three new urban apps in the city of Bilbao is described. The two-phase evaluation methodology designed and the evaluation results of pre-pilot sub-phase are also presented. As a result, an early user experience evaluation for WeLive has been obtained.

Violence continues being an important problem in the society. Thousands of people suffer its effects every day and statistics show this number has maintained or almost increased recently. In the modern environment of smart cities there is a necessity to develop a system capable of detecting if a violent situation is taking place or not. In this paper we present an automatic acoustic violence detection system for smart cities, integrating both signal processing and pattern recognition techniques. The proposed software has been implemented in three steps: feature extraction in time and frequency domain, genetic algorithm implementation in order to select the best features, and classification to take a binary decision. Results derived from the experiments show that MFCCs are the best features for violence detection, and others like pitch or short time energy have also a good performance. In other words, features that can distinguish between voiced and unvoiced frames seem to be a good election for violence detection in real environments.

Nowadays, most of event halls are prepared with normal wireless microphone system that depends mainly on FM radio frequencies. These systems have unexpected problems such as no enough microphones, interferences, weak signal, running out of battery life, unwanted noise, and feedback. Moreover, extra equipment is required if the speech needs recording.The proposed solution is “MyMic”. It is a mobile application developed by ALTARIQ Systems & Projects to solve the previous issues. The main idea of the application is using the smartphone connected to WiFi as a wireless mic. A control panel running on PC/Tablet (server) manages Mobiles (Clients); it receives the sound streams using UDP packets and sends processed sound stream to the sound system. The control panel allows multiple speakers simultaneously. System testing inside a hotel event hall was successful with clear sound of three simultaneous speaking users where noise values were 4.7 % in average.

Multi-sensory monitoring has developed significantly over the last decade. The main reason for this is because of two basic, social needs: security and health. Without doubt, there is a growing social need for security, for detecting dangerous situations, and criminal behavior detection (surveillance). We propose design a semantic framework to representing the human behavior in surveillance context. The principal goal, is have an easy semantic tool that allowing for knowledge engineers get semantic representations case study easy to analyze and reuse. We show examples of how easily and reuse is possible, however we are working on improving our framework, in order to make it much more manageable for the semantic use. Conclusions and future work of our research were generated to visualize what the new tools to be developed.

In this paper a secure distributed system is proposed to manage patients’ information in an emergency service in order to improve efficiency. Authentication, confidentiality and automatic and robust patient identification are provided. The system elements are NFC wristbands, assigned to patients, mobile devices assigned to medical staff and a server to manage the information and verify that its use is legitimate. Patient’s identification is carried out through a keyed-Hash Message Authentication Code. In addition, a challenge-response protocol is proposed for mutual authentication of medical staff and the server using the FullIdent scheme Identity Based Encryption. The same scheme gives confidentiality to the system. The definition of this system provides a secure solution based on mHealth for managing patients in an emergency service.

Domestic violence is one of the most serious and widespread problems in our society. In the most dangerous cases, the use of special devices for GPS tracing is recommended. However, the truth is that they are used only in extreme cases. This work describes the idea and operation of a mobile application to combat domestic violence. Among the main technologies that are used in the implementation of the proposal are Android, BLE and LTE. If an offender gets close to a victim, even if the established threshold distance has not been broken, the victim is warned thanks to the application. Besides, if the threshold distance is broken, automatic streaming of a real-time video is sent to the police and a list of stored contacts are warned so that they can try to help to protect the victim till the police arrives.

Earthquakes have a profound social and economic impact when they hit densely populated urban areas. Early and Quick Safety and Structural Stability Assessment of Critical Physical Infrastructure, affected by large-scale disasters such as earthquakes, is a key task that determines potential outcomes of further activities like urban search and rescue, and recovery of the urban area. This task needs to be conducted as efficient, effective, and consistent as possible. This article presents a preliminary and exploratory effort to understand the role that technology may play in improving assessment of critical physical infrastructure affected by disasters, through a video game-based simulator. Preliminary results include validation from subject matter experts from the Chilean Department of Transportation. The results of this work are encouraging for the development of a national standard capability on conducting assessment of critical physical infrastructure affected by disasters.

Public-key cryptography is too slow for general purpose encryption, with most applications limiting its use as much as possible. Some secure protocols, especially those that enable forward secrecy, make a much heavier use of public-key cryptography, increasing the demand for lightweight cryptosystems that can be implemented in low powered or mobile devices. This performance requirements are even more significant in critical infrastructure and emergency scenarios where peer-to-peer networks are deployed for increased availability and resiliency. We benchmark several public-key key-exchange algorithms, determining those that are better for the requirements of critical infrastructure and emergency applications.

The adoption of RFID (Radio Frequency Identification) technology has greatly improved the efficiency of inventory control, supply-chain management and logistics. With RFID group scanning, an RFID reader scans a group of RFID tagged objects to generate a grouping proof of “simultaneous” presence. Shipments may have to be tracked remotely by readers that are not necessarily trusted. In this paper we extend earlier work on grouping proofs and group codes to capture resilient group scanning with untrusted readers. We present a two-pass anonymous grouping proof (of integrity) for the scanned objects of a (not necessarily complete) collection of RFID tagged objects that identifies missing objects. The proof is generated by an untrusted reader and can be checked by a verifier, a trusted entity. We only assume that RFID tags are able to generate pseudorandom numbers and compute one-way hash functions.