Worldwide, one in three people do not have access to safe drinking water, and two out of five people do not have a basic facility for washing hands with soap and water, according to goal 6 Sustainable Development of United Nations, clean water and sanitation. Due to different meteorological factors that are happening today altering the water cycle and therefore making its regeneration more difficult. The use of fog and obtaining atmospheric water becomes crucial for low-income communities where their supply is limited for their needs. The objectives of this work were to describe the seven fog collector systems that have been installed in various areas of Ecuador, as well as the implementation of IoT sensors to measure some climatic variables in conjunction with the capacity to capture water from the fog, finally, make correlation and covariance matrices of the data generated by the sensors. The fog collector system that worked best was that of Galte Laime with an average water intake of 1.91 L/m2/day. The correlation matrix shows strong positive relationships among temperature, environmental light, and battery.

MOONS (Multi-Object Optical and Near-infrared Spectrograph) is a third-generation visible and near-infrared spectrograph for the ESO Very Large Telescope, currently nearing the end of the assembly phase. The three channel spectrograph is fed via a fibre positioning module (FPM) which configures the location of 1001 fibres. The robotic fibre positioning units (FPUs) have been jointly developed by the UK Astronomy Technology Centre (UKATC) and MPS Microsystems (MPS) and provide a high-performance multiplexed focal plane with excellent transmission characteristics. An overview of the as-built mechanisms and supporting infrastructure is presented, with details on the extensive calibration process carried out. The integration process to date will be described, including a discussion of key lessons learned.

Environmental monitoring is critical to develop appropriate policies for environmental sustainability. The quality of drinking and agricultural water is affected by bad environmental management strategies. In addition, natural water sources have been reduced due to climate changes and the lack of police and compliance with regulations for water resources management. In the Andes region of Ecuador, these problems have increased in recent decades. The lack of water affects not only human well-being, but also causes soil degradation. Therefore, finding alternative water sources has become the underlying requirement to provide human well-being and mitigate soil degradation in these regions. This article presents a fog collection system based on water condensation towers and Internet of Things (IoT) technology with real-time monitoring. Our system allows monitoring of environmental parameters while providing alternative sources of water from the environmental fog. Therefore, by collecting information related to the process of fog collection and climatic measurements of the environment, we seek to determine the state and trends of environmental conditions with respect to the performance of fog collection. In addition, our system allows the storage of historical environmental data, which can be used to develop environmental management policies. We have deployed our system on the slopes of the Ilalo volcano, Pichincha province, being part of the Ecuadorian Andes, where the soil deterioration has increased in recent years and has the largest soil degradation rate in Ecuador

MOONS (Multi-Object Optical and Near-infrared Spectrograph) is a third-generation visible and near-infrared spectrograph for the ESO Very Large Telescope currently under construction. The instrument's spectroscopic capabilities are multiplexed via a fibre positioning module (FPM) which configures the location of 1001 fibres. The fibre positioning units (FPUs) have been jointly developed by the UK Astronomy Technology Centre (UKATC) and MPS Microsystems (MPS) to optimise instrument efficiency by providing excellent transmission and an open-loop positioning strategy, allowing a tightly packed focal plane to be rapidly reconfigured. The mechanism geometry enables all positions in the focal plane to be observed in conjunction with a companion sky fibre at close separation. A description of the as-manufactured design and production process of the FPUs is presented, along with a discussion of the performance proven to date, including achievement of the critical pupil alignment and positional repeatability requirements. An overview of the custom testing rig built to automate the characterisation and calibration process is also presented.

Many indoor localization solutions rely on common radio signal strength indication or on Inertial Measurement Units (IMU). In this work, we present a server-based indoor positioning algorithm based on time of flight measurements of ultra-wideband (UWB) signals for ranging, IMUs for movement detection, and floor plan information. We implemented a particle filter to fuse all the information to achieve high indoor localization performance. We evaluated our system, running on a centralized server connected to the target and anchor Raspberry Pi devices equipped with Sequitur Pi UWB transmitters, in complex real indoor environments. Moreover, we compared our system to the commercial indoor localization system Sequitur InGPS Lite, distributed by UNISET. Results show that our algorithm could achieve an average tracking error of 0.45m and a 90% accuracy of 0.87m. Thus, our prototype can keep up with the Sequitur InGPS Lite system and outperforms previous signal strength implementations, which makes it highly promising for future research.

Indoor location awareness enables many location-based services, such as smart homes or smart offices. The huge amount of sensor data collected by nowadays’ smartphones provides a solid basis for applying advanced machine learning algorithms to derive the correlation between indoor locations and sensor measurements. The combination of multiple sensor measurements, such as the Received Signal Strength of surrounding Wi-Fi access points and magnetic fields, is assumed to be unique in many locations, which can be derived to accurately predict smartphones’ indoor locations. In this work, we propose a novel ensemble learning method to provide room level indoor localization in smart buildings. The proposal is based on a conditional probability model, which combines prediction results of multiple individual machine learning predictors using conditional probability concepts to predict class labels. We have implemented the system on Android smartphones and conducted extensive experiments in real-world office-like environments. The experiment results show that the proposed ensemble predictor outperforms individual and ensemble voting-based machine learning algorithms. It achieves the best indoor landmark localization accuracy of nearly 97% in office-like environments. This work provides a coarse-grained indoor room recognition, which can be envisioned as a basis for accurate indoor positioning.

Seamless outdoor-indoor positioning plays a critical role in many emerging applications, e.g., large-coverage user navigation in cities, smart buildings, and analytics of user spatial location big data. It is still challenging to construct a large-scale seamless outdoor-indoor positioning system due to the limited coverage of indoor positioning. In this paper, we propose a seamless outdoor-indoor crowdsensing positioning (SoiCP) system in which a radio map is automatically constructed based on crowdsourcing pedestrian dead reckoning (PDR) traces without professional site surveying. The constructed radio map is robust to inaccurate PDR traces and does not rely on prior knowledge of floor plans. In SoiCP, the crowdsensed radio map is obtained by a proposed three-step trace matching algorithm. This algorithm leverages building gates and WiFi fingerprints as landmarks to merge the noisy crowdsourcing traces and accurately construct the user walking paths. Moreover, following the crowdsensed radio map, SoiCP uses an enhanced particle filter to fuse PDR, GPS, and WiFi fingerprinting for seamless outdoor-indoor positioning with high accuracy. The comprehensive real-world experiments in two large-scale shopping malls demonstrate that SoiCP can effectively crowdsense the walking paths and track moving users with high accuracy.

Positioning is envisioned as an essential enabler of future fifth generation (5G) mobile networks due to the massive number of use cases that would benefit from knowing users' positions. In this work, we propose a particle filter-based reinforcement learning (PFRL) approach for the robust wireless indoor positioning system. Our algorithm integrates information of indoor zone prediction, inertial measurement units, wireless radio-based ranging, and floor plan into an particle filter. The zone prediction method is designed with an ensemble learning algorithm by integrating individual discriminative learning methods and Hidden Markov Models. Further, we integrate the particle filter approach with a reinforcement learning-based resampling method to provide robustness against localization failure problems such as the kidnapping robot problem. The PFRL approach is validated on a two-tier architecture, in which distributed machine learning tasks are hosted at client and edge layer. Experiment results show that our system outperforms traditional terminal-based approaches in both stability and accuracy.

Software-Defined Networking (SDN) is a promising approach to simplify the management of Wireless Sensor Networks (WSNs). Many SDN frameworks for WSNs have been proposed, while real-world testbeds to accelerate the development of SDN-based WSN applications are still rare. In this work, we propose SDNWisebed: an SDN-empowered WSN testbed management system that enhances the WSN management functions with a stateful Software-Defined Networking solution. This testbed was designed to evaluate various types of SDN-based WSN applications and enhance their performance, such as WSN routing protocols and network applications, before deploying them in real-world infrastructures. To validate its efficiency, we conducted both functional and performance evaluation. Real-world experiment results show that the speed of integration of new SDN applications can be improved thanks to the stateful feature awareness of SDNWisebed.

Real-time localization is the underlying requirement for providing context-aware services in the Internet of Things (IoT), Although several methods have been proposed to provide indoor localization, most of them implement the running algorithms locally in the mobile device to be located. However, the limited computational resources of mobile devices make it difficult to run complex algorithms. As an alternative, Multi-Access Edge Computing (MEC) as a promising paradigm extends the traditional cloud computing capabilities towards the edge of the network. This enables accurate location-aware services. In this work, we present an indoor tracking system based on the MEC paradigm for ultra wide band devices. Our tracking algorithms fuse machine learning-based zone prediction, Ultra Wide Band (UWB) radio ranging, inertial measurement units, and floor plan information into an enhanced particle filter. The localization process is hosted in an Edge server, which performs the resource-demanding calculation that is offloaded from the client devices. Moreover, the client devices are also equipped with certain processing power to handle sensor data processing. Our system includes also a Cloud layer, which enables data storage and data visualization for multiple clients. We evaluate our system in two complex environments. Experiment results show that our tracking system can achieve the average tracking error of 0.49 meters and 90% accuracy of 0.6 meters in real-time.

Nowadays, smartphones can collect huge amounts of data from their surroundings with the help of highly accurate sensors. Since the combination of the Received Signal Strengths of surrounding access points and sensor data is assumed to be unique in some locations, it is possible to use this information to accurately predict smartphones' indoor locations. In this work, we apply machine learning methods to derive the correlation between smartphones' locations and the received Wi-Fi signal strength and sensor values. We have developed an Android application that is able to distinguish between rooms on a floor, and special landmarks within the detected room. Our real-world experiment results show that the Voting ensemble predictor outperforms individual machine learning algorithms and it achieves the best indoor landmark localization accuracy of 94% in office-like environments. This work provides a coarse-grained indoor room recognition and landmark localization within rooms, which can be envisioned as a basis for accurate indoor positioning.

Due to the growing area of ubiquitous mobile applications, indoor localization of smartphones has become an interesting research topic. Most of the current indoor localization systems rely on intensive site survey to achieve high accuracy. In this work, we propose an efficient smartphones indoor localization system that is able to reduce the site survey effort while still achieving high localization accuracy. Our system is built by fusing a variety of signals, such as Wi-Fi received signal strength indicator, magnetic field and floor plan information in an enhanced particle filter. To achieve high and stable performance, we first apply discriminative learning models to integrate Wi-Fi and magnetic field readings to achieve room level landmark detection. Further, we integrate landmark detection, range-based localization models, with a graph-based discretized system state representation. Because our approach requires only discriminative learning-based room level landmark detections, the time spent in the learning phase is significantly reduced compared to traditional Wi-Fi fingerprinting or landmark-based approaches. We conduct experimental studies to evaluate our system in an office-like indoor environment. Experiment results show that our system can significantly reduce the learning efforts, and the localization method can achieve performance with an average localization error of 1.55 meters.

Smartphones are a key enabling technology in the Internet of Things (IoT) for gathering crowd-sensed data. However, collecting crowd-sensed data for research is not simple. Issues related to device heterogeneity, security, and privacy have prevented the rise of crowd-sensing platforms for scientific data collection. For this reason, we implemented VIVO, an open framework for gathering crowd-sensed Big Data for IoT services, where security and privacy are managed within the framework. VIVO introduces the enrolled crowd-sensing model, which allows the deployment of multiple simultaneous experiments on the mobile phones of volunteers. The collected data can be accessed both at the end of the experiment, as in traditional testbeds, as well as in real-time, as required by many Big Data applications. We present here the VIVO architecture, highlighting its advantages over existing solutions, and four relevant real-world applications running on top of VIVO.

Location aware services in the Internet of Things are essential for smart environments. Location awareness enables operational systems to deliver useful information for supplying context-aware applications. We propose an efficient probabilistic model to provide good and stable localization accuracy in smart building environments for smartphones. Our proposed localization method fuses zone detection, radio-based ranging, inertial measurement units and floor plan information into an enhanced particle filter. Zone detection is designed with an ensemble learning algorithm by combining Hidden Markov Models and discriminative learning methods. We first apply ensemble learning models to achieve zone detection. Further, we integrate zone detection and an enhanced ranging model to achieve high and stable localization performance. Experiment results in an office-like indoor environment show that our system outperforms traditional localization approaches considering stability and accuracy. The localization method can achieve performance with an average localization error of 1.26 meters.

An accurate room localization system is a powerful tool for providing location-based services. Considering that people spend most of their time indoors, indoor localization systems are becoming increasingly important in designing smart environments. In this work, we propose an efficient ensemble learning method to provide room level localization in smart buildings. Our proposed localization method achieves high room-level localization accuracy by combining Hidden Markov Models with simple discriminative learning methods. The localization algorithms are designed for a terminal-based system, which consists of commercial smartphones and Wi-Fi access points. We conduct experimental studies to evaluate our system in an office-like indoor environment. Experiment results show that our system can overcome traditional individual machine learning and ensemble learning approaches.

The rapid growth area of ubiquitous applications and location-based services has made indoor localization an interesting topic for research. Some indoor localization solutions for smartphones exploit radio information and Inertial Measurement Units (IMUs), which are embedded in most of the modern smartphones. In this work, we propose to fuse WiFi Receiving Signal Strength Indicator (RSSI) readings, IMUs, and floor plan information in an enhanced particle filter to achieve high accuracy and stable performance in the tracking process. We provide an efficient double resampling method to mitigate errors caused by off-the-shelf IMUs and WiFi sensors embedded in commodity smartphones. The algorithms are designed in a terminal-based system, which consists of commercial smartphones and WiFi access points. We evaluate our system in two complex environments along moving paths. Experiment results show that our tracking method can achieve the average tracking error of $1.01$ meters and $90\%$ accuracy of $1.7$ meters.

Accurate and reliable smartphone indoor localization is fundamental for indoor location-based services (LBS). Smart environments, such as smart offices, interconnect office facilities, indoor wireless sensor and actuator networks (WSANs), smartphones, and human to provide comfortable user experiences. To smoothly integrate the localization algorithms with WSAN infrastructures, a combination of both hardware and software components is required. In this work, we present a system for creating indoor location-aware smart office environments using wireless sensor and actuator networks. Our system includes a smartphone indoor localization module, a WSAN responsible for environmental monitoring and actuator activation, and a gateway that interconnects WSAN, indoor localization module with smartphone users. To reduce the efforts of data collection, we have designed a semi-supervised learning-based indoor localization mechanism, which uses only a small amount of labeled data and a big amount of unlabeled data. The system is based on data fusion of Wi-Fi RSSI and smartphone onboard IMU readings. We implemented a system prototype and performed intensive experiments in indoor office environments to evaluate the system performance. The system could accurately locate real-time positions of occupants, which could trigger the retrieval of environmental measurements and activate the office appliances automatically (e.g. turn on/off lights) based on the estimated locations and correlated environmental sensor information.

Smart environments interconnect indoor building environments, indoor wireless sensor and actuator networks, smartphones, and human together to provide smart infrastructure management and intelligent user experiences. To enable the "smart" operations, a complete set of hardware and software components are required. In this work, we present Smart Syndesi, a system for creating indoor location-aware smart building environments using wireless sensor and actuator networks (WSANs). Smart Syndesi includes an indoor tracking system and a WSAN for environmental monitoring and actuator activation, interconnected via a gateway with mobile users. The indoor positioning system tracks the real-time location of occupants with high accuracy, which works as a basis for indoor location-based sensor actuation automation. To show how the multiple software/hardware components can be integrated, we implemented a system prototype and performed intensive experiments in indoor office environments.