Fuzzy min-max neural networks for categorical data: application to missing data imputation

The fuzzy min–max neural network classifier is a supervised learning method. This classifier takes the hybrid neural networks and fuzzy systems approach. All input variables in the network are required to correspond to continuously valued variables, and this can be a significant constraint in many real-world situations where there are not only quantitative but also categorical data. The usual way of dealing with this type of variables is to replace the categorical by numerical values and treat them as if they were continuously valued. But this method, implicitly defines a possibly unsuitable metric for the categories. A number of different procedures have been proposed to tackle the problem. In this article, we present a new method. The procedure extends the fuzzy min–max neural network input to categorical variables by introducing new fuzzy sets, a new operation, and a new architecture. This provides for greater flexibility and wider application. The proposed method is then applied to missing data imputation in voting intention polls. The micro data—the set of the respondents’ individual answers to the questions—of this type of poll are especially suited for evaluating the method since they include a large number of numerical and categorical attributes.

Categorical Missing Data Imputation Using Fuzzy Neural Networks with Numerical and Categorical Inputs

There are many situations where input feature vectors are incomplete and methods to tackle the problem have been studied for a long time. A commonly used procedure is to replace each missing value with an imputation. This paper presents a method to perform categorical missing data imputation from numerical and categorical variables. The imputations are based on Simpson’s fuzzy min-max neural networks where the input variables for learning and classification are just numerical. The proposed method extends the input to categorical variables by introducing new fuzzy sets, a new operation and a new architecture. The procedure is tested and compared with others using opinion poll data.

Sonar bat for visually impaired people

Visually impaired people have many difficulties when traveling because it is impossible for them to detect obstacles that stand in their way. Bats instead of using the sight to detect these obstacles use a method based on ultrasounds, as their sense of hearing is much more developed than that of sight. The aim of the project is to design and build a device based on the method used by the bats to detect obstacles and transmit this information to people with vision problems to improve their skills. The method involves sending ultrasonic waves and analyzing the echoes produced when these waves collide with an obstacle. The sent signals are pulses and the information needed is the time elapsed from we send a pulse to receive the echo produced. The speed of sound is fixed within the same environment, so measuring the time it takes the wave to make the return trip, we can easily know the distance where the object is located. To build the device we have to design the necessary circuits, fabricate printed circuit boards and mount the components. We also have to design a program that would work within the digital part, which will be responsible for performing distance calculations and generate the signals with the information for the user. The circuits are the emitter and the receiver. The transmitter circuit is responsible for generating the signals that we will use. We use an ultrasonic transmitter which operates at 40 kHz so the sent pulses have to be modulated with this frequency. For this we generate a 40 kHz wave with an astable multivibrator formed by NAND gates and a train of pulses with a timer. The signal is the product of these two signals. The circuit of the receiver is a signal conditioner which transforms the signals received by the ultrasonic receiver in square pulses. The received signals have a 40 kHz carrier, low voltage and very different shapes. In the signal conditioner we will amplify the voltage to appropriate levels, eliminate the component of 40 kHz and make the shape of the pulses square to use them digitally. To simplify the design and manufacturing process in the digital part of the device we will use the Arduino platform. The pulses sent and received echoes enter through input pins with suitable voltage levels. In the Arduino, our program will poll these two signals storing the time when a pulse occurs. These time values are analyzed and used to generate an audible signal with the user information. This information is stored in the frequency of the signal, so that the generated signal frequency varies depending on the distance at which the objects are. RESUMEN Las personas con discapacidad visual tienen muchas dificultades a la hora de desplazarse ya que les es imposible poder detectar los obstáculos que se interpongan en su camino. Los murciélagos en vez de usar la vista para detectar estos obstáculos utilizan un método basado en ultrasonidos, ya que su sentido del oído está mucho más desarrollado que el de la vista. El objetivo del proyecto es diseñar y construir un dispositivo basado en el método usado por los murciélagos para detectar obstáculos y que pueda ser usado por las personas con problemas en la vista para mejorar sus capacidades. El método utilizado consiste en enviar ondas de ultrasonidos y analizar el eco producido cuando estas ondas chocan con algún obstáculo. Las señales enviadas tendrán forma de pulsos y la información necesaria es el tiempo transcurrido entre que enviamos un pulso y recibimos el eco producido. La velocidad del sonido es fija dentro de un mismo entorno, por lo que midiendo el tiempo que tarda la onda en hacer el viaje de ida y vuelta podemos fácilmente conocer la distancia a la que se encuentra el objeto. Para construir el dispositivo tendremos que diseñar los circuitos necesarios, fabricar las placas de circuito impreso y montar los componentes. También deberemos diseñar el programa que funcionara dentro de la parte digital, que será el encargado de realizar los cálculos de la distancia y de generar las señales con la información para el usuario. Los circuitos diseñados corresponden uno al emisor y otro al receptor. El circuito emisor es el encargado de generar las señales que vamos a emitir. Vamos a usar un emisor de ultrasonidos que funciona a 40 kHz por lo que los pulsos que enviemos van a tener que estar modulados con esta frecuencia. Para ello generamos una onda de 40 kHz mediante un multivibrador aestable formado por puertas NAND y un tren de pulsos con un timer. La señal enviada es el producto de estas dos señales. El circuito de la parte del receptor es un acondicionador de señal que transforma las señales recibidas por el receptor de ultrasonidos en pulsos cuadrados. Las señales recibidas tienen una portadora de 40 kHz para poder usarlas con el receptor de ultrasonidos, bajo voltaje y formas muy diversas. En el acondicionador de señal amplificaremos el voltaje a niveles adecuados además de eliminar la componente de 40 kHz y conseguir pulsos cuadrados que podamos usar de forma digital. Para simplificar el proceso de diseño y fabricación en la parte digital del dispositivo usaremos la plataforma Arduino. Las señales correspondientes el envío de los pulsos y a la recepción de los ecos entraran por pines de entrada después de haber adaptado los niveles de voltaje. En el Arduino, nuestro programa sondeara estas dos señales almacenando el tiempo en el que se produce un pulso. Estos valores de tiempo se analizan y se usan para generar una señal audible con la información para el usuario. Esta información ira almacenada en la frecuencia de la señal, por lo que la señal generada variará su frecuencia en función de la distancia a la que se encuentren los objetos.

Localization and Tracking in Energy-Constrained Networks with Dynamic Connectivity

Systems used for target localization, such as goods, individuals, or animals, commonly rely on operational means to meet the final application demands. However, what would happen if some means were powered up randomly by harvesting systems? And what if those devices not randomly powered had their duty cycles restricted? Under what conditions would such an operation be tolerable in localization services? What if the references provided by nodes in a tracking problem were distorted? Moreover, there is an underlying topic common to the previous questions regarding the transfer of conceptual models to reality in field tests: what challenges are faced upon deploying a localization network that integrates energy harvesting modules? The application scenario of the system studied is a traditional herding environment of semi domesticated reindeer (Rangifer tarandus tarandus) in northern Scandinavia. In these conditions, information on approximate locations of reindeer is as important as environmental preservation. Herders also need cost-effective devices capable of operating unattended in, sometimes, extreme weather conditions. The analyses developed are worthy not only for the specific application environment presented, but also because they may serve as an approach to performance of navigation systems in absence of reasonably accurate references like the ones of the Global Positioning System (GPS). A number of energy-harvesting solutions, like thermal and radio-frequency harvesting, do not commonly provide power beyond one milliwatt. When they do, battery buffers may be needed (as it happens with solar energy) which may raise costs and make systems more dependent on environmental temperatures. In general, given our problem, a harvesting system is needed that be capable of providing energy bursts of, at least, some milliwatts. Many works on localization problems assume that devices have certain capabilities to determine unknown locations based on range-based techniques or fingerprinting which cannot be assumed in the approach considered herein. The system presented is akin to range-free techniques, but goes to the extent of considering very low node densities: most range-free techniques are, therefore, not applicable. Animal localization, in particular, uses to be supported by accurate devices such as GPS collars which deplete batteries in, maximum, a few days. Such short-life solutions are not particularly desirable in the framework considered. In tracking, the challenge may times addressed aims at attaining high precision levels from complex reliable hardware and thorough processing techniques. One of the challenges in this Thesis is the use of equipment with just part of its facilities in permanent operation, which may yield high input noise levels in the form of distorted reference points. The solution presented integrates a kinetic harvesting module in some nodes which are expected to be a majority in the network. These modules are capable of providing power bursts of some milliwatts which suffice to meet node energy demands. The usage of harvesting modules in the aforementioned conditions makes the system less dependent on environmental temperatures as no batteries are used in nodes with harvesters--it may be also an advantage in economic terms. There is a second kind of nodes. They are battery powered (without kinetic energy harvesters), and are, therefore, dependent on temperature and battery replacements. In addition, their operation is constrained by duty cycles in order to extend node lifetime and, consequently, their autonomy. There is, in turn, a third type of nodes (hotspots) which can be static or mobile. They are also battery-powered, and are used to retrieve information from the network so that it is presented to users. The system operational chain starts at the kinetic-powered nodes broadcasting their own identifier. If an identifier is received at a battery-powered node, the latter stores it for its records. Later, as the recording node meets a hotspot, its full record of detections is transferred to the hotspot. Every detection registry comprises, at least, a node identifier and the position read from its GPS module by the battery-operated node previously to detection. The characteristics of the system presented make the aforementioned operation own certain particularities which are also studied. First, identifier transmissions are random as they depend on movements at kinetic modules--reindeer movements in our application. Not every movement suffices since it must overcome a certain energy threshold. Second, identifier transmissions may not be heard unless there is a battery-powered node in the surroundings. Third, battery-powered nodes do not poll continuously their GPS module, hence localization errors rise even more. Let's recall at this point that such behavior is tight to the aforementioned power saving policies to extend node lifetime. Last, some time is elapsed between the instant an identifier random transmission is detected and the moment the user is aware of such a detection: it takes some time to find a hotspot. Tracking is posed as a problem of a single kinetically-powered target and a population of battery-operated nodes with higher densities than before in localization. Since the latter provide their approximate positions as reference locations, the study is again focused on assessing the impact of such distorted references on performance. Unlike in localization, distance-estimation capabilities based on signal parameters are assumed in this problem. Three variants of the Kalman filter family are applied in this context: the regular Kalman filter, the alpha-beta filter, and the unscented Kalman filter. The study enclosed hereafter comprises both field tests and simulations. Field tests were used mainly to assess the challenges related to power supply and operation in extreme conditions as well as to model nodes and some aspects of their operation in the application scenario. These models are the basics of the simulations developed later. The overall system performance is analyzed according to three metrics: number of detections per kinetic node, accuracy, and latency. The links between these metrics and the operational conditions are also discussed and characterized statistically. Subsequently, such statistical characterization is used to forecast performance figures given specific operational parameters. In tracking, also studied via simulations, nonlinear relationships are found between accuracy and duty cycles and cluster sizes of battery-operated nodes. The solution presented may be more complex in terms of network structure than existing solutions based on GPS collars. However, its main gain lies on taking advantage of users' error tolerance to reduce costs and become more environmentally friendly by diminishing the potential amount of batteries that can be lost. Whether it is applicable or not depends ultimately on the conditions and requirements imposed by users' needs and operational environments, which is, as it has been explained, one of the topics of this Thesis.

Herramienta de certificación para la Bioconstrucción

En este trabajo se presenta una propuesta de herramienta para la certificación de viviendas de Bioconstrucción en España. Se basa en que la Bioconstrucción es una práctica consolidada en el territorio Español, pero que carece de estudio científico para definir, caracterizar y comprobar esta práctica como alternativa constructiva comprometida con el medio ambiente y con la salud humana. Primeramente se hace el estudio de las referencias específicas existentes y se complementa con bibliografía afín, para así tener una caracterización en la que pueda asentarse el trabajo de campo. En el trabajo de campo se comprueban los aspectos que componen la Bioconstrucción. Se han buscado casos de estudio diversificados y una detallada evaluación de proyectos y construcciones, así como de la información obtenida en encuestas a los propietarios y usuarios, y a los profesionales responsables de los procesos constructivos. Con esta base de datos se establecieron características tanto constructivas como geográficas o socio ambientales, y se ha definido la estructura para la herramienta. Se propone entonces una metodología de certificación que integra tres dimensiones identificadas a lo largo del estudio de la Bioconstrucción: salud, medio ambiente y responsabilidad social. A partir de una visión sistémica y basándose en valores como la promoción de espacios de la salud humana en consonancia con los patrones de la naturaleza, el equilibrio ambiental hacia al impacto positivo, y la acción consciente de respecto, dignidad y equidad social durante todo su ciclo de vida; con el objetivo de poder evaluarla en todos sus aspectos y contribuir al avance de las políticas de habitabilidad, medioambiental y de responsabilidad social. Se establecen 10 apartados fundamentales: territorio, bioclimática, saludabilidad, materiales, sistemas constructivos, energías, agua, aire, residuos y gestión. Cada uno se desarrolla en forma de requerimientos, en un total de 175, que evalúan los diversos aspectos de una vivienda de Bioconstrucción. En la valoración cada requerimiento es evaluado en su relación con cada etapa del ciclo de vida de la edificación: concepción, construcción, ocupación y pos-ocupación. Así la valoración será transversal y multidisciplinar, permitiendo análisis tanto parciales como globales para que se pueda mejorar el tema que no cumple con el mínimo pedido. El resultado de la evaluación es la valoración del estado “bio” de la construcción o proyecto. Se hace la validación de la herramienta en casos estudiados. Los resultados de la investigación son la definición y la caracterización de la Bioconstrucción en España, una base de datos de más de 100 casos de estudio, una herramienta específica y validada que evidencia diversas líneas de investigación dónde profundizar y fomentar su desarrollo. Con ello, se pretende contribuir al cambio de paradigma que hace de la construcción actual un desafío hacia un contexto equilibradamente respetuoso con toda la vida que habita nuestro ecosistema. ABSTRACT This work presents a certification tool for the Bioconstrucción's housings in Spain. Based in that the Bioconstrucción is a consolidated practice in the Spanish territory, but there ins´t scientific study to define, to characterize and to verify this practice as a constructive alternative compromised with the environment and with the human health. At first we study about the specific references and them complements itself with related bibliography. By this way to have a characterization in the one that could settle the fieldwork. In the fieldwork there are verified the aspects that compose the Bioconstrucción. Cases of study have been looked diversified and a detailed evaluation of projects and constructions, as well as of the information obtained in you poll the owners and users, and to the professionals responsible for the constructive processes. With this database there were established both constructive and geographical characteristics or partner environmental, and the structure has been defined for the tool. The certification methodology proposes integrates three dimensions identified along the study of the Bioconstrucción: health, environment and social responsibility. From a systemic vision and basing in values as the promotion spaces of the human health in agreement with the bosses of the nature, the environmental balance towards to the positive impact, and the conscious action of respect, dignity and social equity during all his life cycle. With the aim to be able to evaluate it in all his aspects and contribute to the advance of the policies of habitability, environmental and of social responsibility. Ten fundamental paragraphs are established: territory, bioclimátic, saludability, materials, constructive systems, energies, water, air, residues and management. Each one develops requirements, in a total of 175, who evaluate the diverse aspects of Bioconstrucción's housing. In the valuation every requirement is evaluated in his relation by every stage of the life cycle of the building: conception, construction, occupation and post-occupation. This way we get a transverse and multidisciplinary valuation, allowing both partial and global analysis in order that it could improve the topic that does not expire with the asked minimum. The result of the evaluation is the "bio" condition of the construction or project. The validation of the tool is done in studied cases. The results of the investigation are the definition and the characterization of the Bioconstrucción in Spain, a database of more than 100 cases of study, a specific and validated tool, and diverse lines for new investigation and to promote his development. It is tried to contribute to change a paradigm that does of the current construction a challenge towards a context in a balanced way respectfully with the whole life that lives in our ecosystem.