Se desarrolla un procedimiento encargado de detectar vehículos circulando por carriles en una avenida. Se utiliza el programa YOLO como sensor de objetos. Mediante la integración del YOLO y un programa de visión artificial se facilita el análisis de información en un área dentro de un fotograma. Se dividió la avenida en tres carriles denominados: carril izquierdo (CI), carril central (CC) y carril derecho (CD). Se establece en el fotograma un área rectangular de base y altura conocida. Seguidamente, el YOLO detecta todos los objetos en el fotograma. La información recolectada se agrupa en una lista junto a sus coordenadas. Mediante un algoritmo, se analiza dicha lista para detectar los vehículos pertenecientes al área de trabajo. Los límites de cada carril CI, CC y CD están comprendidos sobre la base del área rectangular. Las posiciones, pertenecientes a la lista de automóviles, se comparan con los límites de los carriles CC, CI y CD. El resultado de las comparaciones permite reconocer vehículos pertenecientes a los carriles CC, CI y CD dentro de una región del fotograma.

As network services and IoT technologies rapidly evolve, in literature there are many anomalies detection proposals based on datasets to deal with cybersecurity threats. Most of this proposal uses structured data classification and they can recognize with a certain degree of accuracy whether a type of traffic is "anomalous" or not. Even what kind of anomaly it has. Nevertheless, previous works do not clearly indicate the technical methodology to set up the data gathered scenarios. As a main contribution, we are going to show a detailed deployment IoT traffic monitor ready for intelligent network traffic classification. Monitoring and sniffers are an essential concept in network management as it helps network operators to determine the network behavior and status of its components. Anomaly detection also depends on monitoring for decision-making. Thus, this paper will describe the creation of a portable network traffic monitor for IoT using Docker container and bridge networking with SDN.

Nowadays, the proliferation of wearable devices has enabled monitoring user behaviours and activities in a non-invasive, autonomous and straightforward way. Moreover, new trend analysis has been boosted by biosignal sensors from wearable trackers, such as inertial or heart rate sensors. The knowledge of such user activity presents a personalized monitoring to prevent any kind of physical or neurological disorders through the sensor evaluation by an expert. To this end, the definition of key indicators and the display of results and relevant analyses require of agile and effective tools. Therefore, this proposal presents a novel web application where the data obtained from a Fitbit Ionic smartwatch wearable are collected, synchronized and compiled to present a summary of an user's daily activity, which is defined by a linguistic description using fuzzy logic to represent the most relevant Health Key Indicators (HKI). Moreover, an analysis of the user's behaviour over time is proposed by inferring relevant patterns from a fuzzy clustering algorithm.

The proliferation of low-cost wearable trackers are allowing users to collect daily data from human activity in a non-invasive way and outside of laboratory environments. Exploiting these data properly enable the supervision and counseling from experts remotely; however, extracting key indicators from the long datastreams is hard, often based on statistical metrics or clustering from raw data which lack interpretability. To solve it, we propose an interpretable definition of key indicators by means of linguistic protoforms which include fuzzy temporal processing and fuzzy semantic quantification. Moreover, we use the protoforms defined by experts to evaluate the source datastream in order to provide a straightforward description of the daily activity of users. Finally, the degrees of truth of each protoform are analyzed using a fuzzy clustering method to provide an interpretable description of the longterm user activity. This work includes a case study where data from a user activity (heart beats per minute and sleep stages) have been collected by a Fitbit wearable device and evaluated by the proposed methodology.

The emerging technology of SDN (Software Defined Networks) separates the data control plane from the forwarding plane while maintaining a centralized control of the network management. The SDN features require new traffic engineering techniques that exploit the global (centralized) network view, and the status and features of traffic flows. Our purpose is to perform traffic engineering in an SDN architecture, using the OpenFlow protocol capabilities and the potential of the SDN controller to collect operational data of the entire network, such as topology, latency, buffer utilization and frame sizes of the controlled devices in order to implement QoS. Considering that the performance of the network is an essential component of Quality of Service (QoS), and congestion is the main factor that affects it, this paper explores a method to search for alternative paths based on data from switches using the Random Early Detection (RED) congestion control mechanism.

Las necesidades de gestionar diversas condiciones ambientales en las ciudades impulsan a que se tomen mediciones en tiempo real en los reservorios de agua para ser analizada por expertos y tomar decisiones al respecto. Para ello se analizaron diferentes sensores y medios de comunicación arribando a una propuesta que cubra tales necesidades. Así, este trabajo propone un conjunto de tecnologías relacionadas a smartcities que ayudan a mantener mediciones de diversas variables del agua. Se muestra además el desarrollo del framework realizado y la interacción de los componentes electrónicos y de telecomunicaciones LoRaWAN que conforman el sistema propuesto. Para finalizar se expone un ejemplo ilustrativo de los datos censados.

El objetivo principal de este artículo es presentar una metodología sencilla para la simulación de una SDN (Software Defined Networking) [1], configurada con el protocolo OpenFlow [2] controladores POX [3] y ODL [4], utilizando la herramienta de simulación Mininet [5], y realizar pruebas de performance mediante la utilización de IPERFv3 [6], mostrando los resultados con el software GNUPlot [7], todos ellos con licencia GNU. Además, se describen los requerimientos y recomendaciones fundamentales para dicha emulación, y se presentan los resultados de pruebas sencillas con el fin de verificar la conectividad, transferencia de datos y operación sobre una topología de prueba. Se orienta al desarrollo de modelos de evaluación de performance que ayuden a los administradores de red a tomar decisiones en base a atributos críticos (tipos de tráfico, servicios, usuarios, etc.) identificados previamente.

In daily life, people have the need to know the identity of a visitor who comes to their homes, regardless of whether they are there at that time. This need is even greater for people who suffer from some kind of disability that prevents them from meeting the visitor. To provide a solution in this sense, this paper proposes a smart model that performs the task of a doorbell, which should recognize the visitor and alert the user. To achieve that, this proposal incorporates technologies for facial recognition of people, notifications to users and management of their responses. The process to solve the problem was divided into interrelated stages and standardization issues are discussed later. Finally, to test the effectiveness of the model, three scenarios were simulated; each one was composed by different households over which the recognition of known and unknown individuals was analyzed.

Multi-Criteria Decision Analysis (MCDA) is a usual activity among organisations and decisions related to people's activities. Due to the complexity of considering multiple criteria, to select an alternative is a non-trivial task. From operative levels to managerial ones, MCDA is implemented by using several (formal and informal) techniques. Two useful techniques that help to make a decision are the Analytic Hierarchy Process (AHP) and MCDA models based on Linguistic Information (LI). This work describes a MCDA framework that combines the mentioned techniques in order to provide more confidence in the decision making process. To test the proposed model, framework was used to select the adequate network configuration to improve quality of service (QoS). Finally, the framework's outputs were compared to real experts' opinions obtaining satisfactory results.

When solving a problem, human beings must face situations in which they should choose among different alternatives by means of reasoning and mental processes. Many of these decision problems are under uncertain environments including vague, imprecise and subjective information that is usually modeled by fuzzy linguistic approach. This approach uses linguistic information or natural language words and its relation to mental reasoning processes of the experts when expressing their assessments. In a decision process multiple criteria can be evaluated which involving multiple experts with different degrees of knowledge. Such process can be modeled by using Multi-granular Linguistic Information (MGLI) and Computing with Words (CW) processes to solve the related decision problems. Once decision makers (experts) provided their opinions, it is necessary to combine all these opinions to obtain a single overall result that can be interpreted. An aggregation operator allows accomplishing this objective calculating a global value in different ways. In this paper we study the use of aggregation operators in multi-criteria decision-making processes comparing them and obtaining conclusions about their use in our framework. Furthermore, we propose a new aggregation operator taking into account the criteria importance to evaluate the alternatives, and then an illustrative example shows its outcomes.