José Luis Herrero

On-line classification of pollutants in water using wireless portable electronic noses.

A portable electronic nose with database connection for on-line classification of pollutants in water is presented in this paper. It is a hand-held, lightweight and powered instrument with wireless communications capable of standalone operation. A network of similar devices can be configured for distributed measurements. It uses four resistive microsensors and headspace as sampling method for extracting the volatile compounds from glass vials. The measurement and control program has been developed in LabVIEW using the database connection toolkit to send the sensors data to a server for training and classification with Artificial Neural Networks (ANNs). The use of a server instead of the microprocessor of the e-nose increases the capacity of memory and the computing power of the classifier and allows external users to perform data classification. To address this challenge, this paper also proposes a web-based framework (based on RESTFul web services, Asynchronous JavaScript and XML and JavaScript Object Notation) that allows remote users to train ANNs and request classification values regardless users location and the type of device used. Results show that the proposed prototype can discriminate the samples measured (Blank water, acetone, toluene, ammonia, formaldehyde, hydrogen peroxide, ethanol, benzene, dichloromethane, acetic acid, xylene and dimethylacetamide) with a 94% classification success rate.

A Web-Based Approach for Classifying Environmental Pollutants Using Portable E-nose Devices

Poor air quality is a major public health concern in both developed and developing countries. Continued exposure to environments with high pollutant content is estimated to cause nearly 2.5 million premature deaths per year all over the world, including 370,000 in Europe. The worldwide concern about pollution is reflected in the development and implementation of increasingly stringent policies to limit the wide range of contaminants that contribute to poor air quality. However, the sensing of these contaminants in a comprehensive, reliable, and low-cost manner still remains a major challenge. Electronic noses are intelligent systems that can classify and quantify different gases and odors. In this article, the authors evaluate which classification approaches are most effective for identifying and distinguishing between environmental pollutants, and they describe a novel Web-based software framework that incorporates these classifiers to detect pollutants.

Bluetooth gas sensing module combined with smartphones for air quality monitoring

This study addresses the development of a miniaturized (60 × 60 mm) Wireless Sensing Module (WSM) for environmental application and air quality detection. The proposed prototype has six sensors: one for humidity, one for ambient temperature (SHT21 from Sensirion), and four for gas detection (MiCS-4514, MiCS-5526 and MiCS-5914 from SGX Sensortech). The core of the system is based on a high performance 8-bit microcontroller, model PIC18F46K80, from Microchip. The obtained data values were transmitted to the Smartphone through a Bluetooth communication module and a home-developed Android app. The discrimination capability of the module is tested with 10 volatile organic compounds (acetone, acetic acid, benzene, ethanol, ethyl acetate, ethylbenzene, formaldehyde, toluene, xylene, and dimethylacetamide) and the effect of humidity and drift of the sensors is also studied. Results show that 88.33% and 92.22% success rates in classification stage are obtained using Multilayer Perceptron with BackPropagation Learning algorithm and Radial-Basis based Neural Networks, respectively.

Guaranteeing Coherent Software Systems when Composing Coordinated Components

Latest trends in coordination models and languages suggest that it must be supported the separated specification of and the dynamic change of coordination constraints. However, little attention has been paid to guaranteeing that, the application of a separately specified coordination pattern to a set of encapsulated objects, or changing the coordination constraints of a software system at run-time, will not produce semantic errors. These kinds of errors would produce an unpredictable system and, consequently, a lack of software quality. In this paper, a method of generating formal interpretable specifications for the reproduction of coordinated environments is presented. The benefits provided by this method are: (i) easy specification, verification and detection of inconsistencies when composing coordination and functional components, (ii) easy verification and detection of inconsistencies where coordination policies are changed dynamically and (iii) simulation of coordinated behaviors. The method is based on the use of the formal specification language Maude (as a simulation tool) and a coordination model. Although the paper adopts Coordinated Roles, it is also shown how the method can be adapted to other compositional coordination models.

A REstfull Approach for Classifying Pollutants in Water Using Neural Networks

E-nose systems are composed by different type of sensors with the capacity of detecting almost any compound or combination of compounds of an odor. In this regard, software approaches can provide learning capacities for classifying odor information. The aim of this paper is to propose a novel approach for on-line classification of pollutants in the water. To extract information about different types of pollutants, a hand-held, lightweight and battery powered instrument with wireless communications has been designed. An Artificial Neural Network (ANNs) is also proposed to classify different types of pollutants in the water, and two different types of ANNs (feedforward with backpropagation) Learning algorithm and Radial Basis Function based networks) have been developed. Furthermore, to support online classification services, a novel framework for developing high performance web applications is also proposed. Two different approaches have been integrated in this framework: the component-based approach is applied to increase the reusability and modularity degree, while RESTful web services provides the architectural style to connect with remote resources. According to this proposal, a web-based application has been developed to detect pollutants in the water.

Balancing Interpretability against Accuracy in Fuzzy Modeling by Means of ACO

Ant colony optimization (ACO) techniques have been revealed as an effective way to improve the interpretability of fuzzy models by reformulating an initial model [1,2]. However, despite this reformulation preserves the initial fuzzy rules, new rules can be added in the search for an interpretability enhancement. Thus, differences between the inferences of the initial and final models can arise due to the interaction among the initial rules and their adjacent new rules. This can lead to changes in the accuracy of the initial model. In order to keep the accuracy of the initial fuzzy model, this work proposes to include the difference between the outputs of the initial and final models as an additional criterion to evaluate the optimality of the reformulated model. This will allow to balance the interpretability against the accuracy within the optimization algorithm, even making the interpretability improvement conditional on a strict preservation of the initial accuracy.