CITT 2017
Accepted Papers


Development of prototype suit for modeling of lower limbs using NodeMcu and Matlab
Abstract: In the present project a prototype was developed, that models the lower limbs movement, unlike other commercial devices, this prototype seeks to obtain data through low-cost electronics in order to be replicated by the scientific community interested in this research line, it integrates electronic boards and programming software such as Arduino, NodeMCU, Accelerometers and Matlab. The sensors have been placed on a shield that facilitates their connection. The connection through a bus of all the sensors is presented in a NodeMCU board, through Wi-Fi sends a table of all this data for processed in Matlab. Data shows the main articular positions of the lower limbs of a person, using the homogeneous transformation matrices we obtain a seven-bar model that shows positions and accelerations produced by individual joints, the data obtained by the suit can be contrast with simulations performed in previous works, providing good results, and allowing future developments in topics such as gait analysis, teleoperation and modeling of physical therapies to ensure their effectiveness.
Optimization of recommendation ITU P. 1812-3 for the propagation losses prediction in Digital Terrestrial Television System
Abstract: This work was developed for the propagation characteristics of outdoor environments in the cities of Quito-Ecuador and Caracas-Venezuela, using one of the most innovating propagation models used in the planning and dimensioning (PyD) of Digital Terrestrial Television (DTT), specifically Recommendation ITU-R P.1812-3. For this, measurements of the received electric field level (Quito) and power level received (Caracas) made for conditions of fixed reception, with DTT standards, were used during the processes followed in both countries for the selection of the DTT standard base. For the development of four versions, the Particle Swarm Optimization (PSO) and genetic algorithm (AG) techniques were applied individually to the measurements mentioned above, which were part of the intelligence computational techniques family. The performance of these versions, in terms of RMSE (Root Mean Square Error), in the prediction of propagation losses in the environments where the measurements were carried out, were compared with the performance of the unmodified version of the ITU recommendation -R P.1812-3, resulting all versions of the developed model here with better performance.
Mammogram Classification Schemes by Using Convolutional Neural Networks
Abstract: This work presents the comparison of two schemes of mammogram classification based on convolutional neural networks (CNN). The main difference between these two classification schemes relies on the number of bits per image pixel, the feature extraction techniques and the number of neurons in the fully connected layer. We use 1070 mammograms from the Digital Database for Screening Mammography (DDSM), which are divided into two categories: benign and malignant mammograms.We use CNN for classification by applying the open source library TensorFlow which is configured on the high level library Keras. In order to tune our classification model parameters, we apply random and grid search algorithms, by combining the batch size, the number of layers, the learning rate and three optimizers: Adadelta, RMSProp and SGD. We evaluate the classification algorithm performances through the accuracy and two loss functions: Categorical Cross-Entropy and Mean Squared Error. The model with the best accuracy has 85.00%, and a mean squared error of 15.00%.
IoT Android gateway for monitoring and controlling a Wireless Sensor Network
Abstract: Precision agriculture and the automation of agricultural processes such as irrigation with the use of current technologies, create the need to devel-op IoT gateways for WSN with Bluetooth Low Energy motes that demonstrate their functionality through a testbed. The challenge of developing both the testbed and an application for Android smart phones with gateway functionali-ties to package the sensor data into MQTT messages and send it to remote serv-er or cloud computing and, in turn, be a tool for monitoring on-site of sensors and controlling the actuators present in the mote motivated the development of this work. We carried out tests on the proposed testbed, which it denotes its functionality, and shows there is a fast and stable IoT gateway with low CPU and RAM usage, ready for IoT applications such as Precision Agriculture. The application design is extensible to other OS for smart phones and tablets.
Risk Analysis of Implanted Electronic Devices in Human Beings
Abstract: This paper is a guide of risk analysis of electronic devices that are implanted in humans. The development of science and technology made global changes about dangers that threaten the security of mankind, however, in recent years there had arisen a new field of analysis for these implanted devices, contributing important elements of analysis. In this research the authors tried to collect all the technological aspects concerning for some electronic devices that can be implanted in the body that can be hacked, to identify the possible computer attacks that they could suffer. The authors research different threats to which these electronic devices are exposed and how to find the way to reduce such threats and vulnerabilities. As well as to differentiate the types of attacks, the way that they operate and how they affect its operational function processes denoting an operational impact of these, such as an analysis of the prevention, detection and mitigation of the main vulnerabilities.
Smart mobile application for the prediction of the loss of the signal in telecommunication networks based on meteorological variables
Abstract: The objective of the research project was developing a mobile app to allow the signal loss prediction. The prediction is based in meteorological variables, because it is looking to support the decision-making by the telecommunication specialists. A data Pre-processing was realized by the outlier’s deletion and variables correlations that resulted in a new Data Set. Different data mining classification techniques were analyzed using an optimization approach. The result of this analysis allow to determinate that, the algorithm based on Artificial Neural Networks was the one who have the better accuracy index. It was almost the 100% of accuracy. The project Aim to taking advantage of obtained model, it was represented in the Predictive Model Markup Language (PMML) and processed with java technologies in an app mobile development. The app name is SignalPred, this app predict signal loss through the signal reading of meteorological variables.
Design and construction of self-driving vehicle with renewable and non-polluting energy
Abstract: Renewable energy is a contribution to society, it cares for the environment and helps the operation of machines at low cost. The prototype was designed with low-cost materials and to operate within the Technical University of Babahoyo campus. Our primary objective is to create a solar vehicle that uses clean energy and be capable of moving people with motor disability with the utilization of a voice recognition system. Tests were carried out with a real-size prototype obtaining satisfactory results, reaching speeds up to 8 km/h on flat surfaces with operating times of approximately 60 minutes when there are sunlight and 20 minutes at night.
A Kinect-based Gesture Recognition Approach for the Design of an Interactive Tourism Guide Application
Abstract: Emerging technologies have revolutionized industrial and economic processes in the present days, tourism being no exception. For the modern tourist, access to readily available information about destinations they visit has become a necessity. In order to remain competitive cities must provide their visitors with interactive tools, which allow them to gain insight of their surroundings so that they feel confident enough to explore points of interest. This article proposes the design of an interactive tourism visitor guide that recognizes gesture commands to display significant information regarding shopping centers, restaurants, hotels and landmarks in the city of Guayaquil, Ecuador. To achieve this objective, a Kinect based natural user interface is deployed to allow tourist to control the tourism guide with the movements of their hands to preview images, data and maps of how to reach popular touristic locations in the city. The software is written in C# and integration with Kinect is handled through the use of the Microsoft Kinect Software Development Kit and Windows Presentation Foundation. The interface recognizes two main gestures when browsing the guide: push-to-press and grip-to-pan. It is intended that the tourism guide be installed in strategic spots in the city of Guayaquil to reinforce local and international tourism. Preliminary results demonstrate that the quality of interaction and overall user experience of the application allows for implementations in a real-world setting.
An Agent-Based Model for Game Development
Abstract: In this paper we describe a new agent-based model for games devel-opment. This model has the advantage that it allows both high- and low-level behavior specifications. These methods reduce the gap between abstract specifi-cation and system implementation. The use of the model is demonstrated with a real-world game example. The proposed results have been used for the develop-ment of the GAMESONOMY platform. GAMESONOMY is a visual tool to cre-ate multiuser games (including serious games) in the cloud. Also, there is a large class of systems with reactive components that can benefit from our model; for example, e-learning environments, monitoring and control systems, telecommut-ing services and e-commerce systems.
Analysis of transport logistics costs in supply chain management by applying fuzzy logic
Abstract: Supply chain management is one of the main concerns of the compa-nies, due to the expected conditions of the markets, have become fac-tors generating risk causing, uncertainty in the process of delivery of goods associated with transportation. Several studies suggest that the main components of a delivery correspond to its quality and timeliness. Other works of estimation of commissions give relevance to the appli-cation of tariffs previously established. In this way, the beginning of the use of methodologies focused on the systematization and in the ob-taining of estimates of tariffs by means of transport is described. In this paper we study the possibility of establishing, through a computational analysis, whether the variables: Service Provision, Condition of the goods and Time of Delivery, have any influence on the freight rate, we will do this analysis using techniques developed with fuzzy logic. As a result of this work, it was demonstrated that transport rates are esti-mated more reasonably in terms of the constant improvement in the service. Finally, these results were obtained from MATLAB’s fuzzy logic module.
A Dataset for Facial Recognition in the Visible, Ultra Violet and Infrared Spectra
Abstract: This paper describes the acquisition process and content of a multispectral face database, which can be used to research on face recognition methods dealing with two of the most challenging problems in this area, i.e. partial occlusion and pose variations. Four cameras were synchronized and arranged to simultaneously capture images from visi- ble, thermal, ultraviolet and near-infrared spectra, which had reported promising results for recognizing faces, individually. In order to simu- late pose variations, each subject was asked to look forward, up, down, and to the sides, varying the point of view angle. On the other hand, partial occlusion was generated using sunglasses and a paper sheet. Ad- ditionally, three lighting changes were also included (halogen, natural and infrared). A total of 306 images were acquired by subject and 31 subjects were recruited. So, the whole database is composed of 9486 im- ages, which are now available to other researchers. Preliminary results showed that spectra variations a ect the performance of a deep learn- ing recognition approach. As far as we know, this is the first database of faces including images from those spectra and the other variations simultaneously.
A proposal for migrating SOA applications to Cloud using Model-Driven Development
Abstract: Software applications are currently considered an element essential and indispensable in all business activities. Nevertheless, for their construction and deployment to use all the resources that are available in remote and accessible locations on the network, which leads to inecient operations in development and deployment, and enormous costs in the acquisition of IT equipment. This paper aims to contribute proposing SOA2Cloud, a framework to migrate SOA applications to Cloud environments, following a Model-Driven Software Development approach. An example of an application that shows the feasibility of our approach was developed.
Delimitation and codification of hydrographic units through the use of Geographic Information Systems
Abstract: Geographic Information Systems (GIS) provide new methods and tools for automatic data processing, which are used by organizations to improve process management. The basin is considered as the basic unit of territorial planification, for which the hydrographic units delimitation is the key for water resources administration. The lack of detail in basins geographical information has to be solved, in order to optimize the decision-making process and related tasks. Using the Pfafstetter methodology, the Jubones river basin was subdivided into level 6. For this purpose, we used the level 5 hydrographic units, the SRTM spatial resolution data model of 30m and GIS software. At level 5, 13943, 13944, 13945, 13947, 13948 and 13949 hydrographic units were identified. In accordance with the selected method guidelines, 9 drainage areas were coded for each hydrographic level 5 unit. 54 hydrographic units at level 6, of which 30 correspond to inter-basins and 24 to basins, thus updating the geographical information. The total subdivided area was 312,223.06 ha. It is important to replicate the research in other basins and publish the results using the Spatial Data Infrastructure (SDI) platforms.
Open Source IoT Technology to connect environment monitoring system
Abstract: Internet of Things (IoT) is transforming the life of people in recent years, because of this concept has been developed in new technologies that people use in their homes, public transportation and their jobs. In addition, the need of people to be connected and interact with the different situations that occur around them, through the use of sensors, low cost embedded systems and actuators.The new devices should be connected and require a clear and reliable communication between each other. This paper describes the implementation of an environmental monitoring system based on a network of wireless sensors and IoT open source gateway that allow scalability and integration with different IoT services in the cloud like AWS IoT. Also this paper shows the new technologies to connect a complete IoT system using low cost hardware and open source gateway platforms based on Linux System.
Diagnosis of the Internet and social networks use of Secondary Students the urban center of Canton Cañar, as a means to determine digital vulnerabilities
Abstract: The present article is the result of a research project executed by the Catholic University of Cuenca of Ca~nar Computer Engineering program: \Proposed Controls For Minimizing The Risks Arising From The Internet And Social Networks On High School Students In The Urban Center of Canton Ca~nar". A Diagnosis of the internet and social networks use was performed as a means to determine digital vulnerabilities. Through a theoretical study it was determined which threats are most commonly inherent in utilizing the internet and social networks. Surveys were applied to 40% of enrolled students. Given surveys then indicated the common uses of the internet, social networks, and potential vulnerabilities that could be exploited by malicious people. Results determined that threats with higher risks are related to violations of privacy, which are visible to behavioral tendencies in the adolescents placing them in vulnerable situations susceptible to privacy violations.
Towards a methodology for Knowledge Discovery in Databases for decision making in the University leveling and admission unit
Abstract: At institutions of higher education, data is generated daily. This mas-sive amount of information is stored in different repositories, and it is increas-ingly difficult to locate specific data on which decisions can be made because universities are unaware of processes that allow for the extraction of valuable and reliable information. In this paper, we present a methodology that includes the Knowledge Discovery in Databases (KDD) coupled with the HEFESTO version 2.0 methodology for the construction of Data Warehouses and the use of Data Mining (DM) techniques. By implementing the proposed methodology, problems stemming from a lack of information relating to student placement and admis-sions in the UNAE and New Student Orientation (NSO) departments at the Pol-ytechnic School of Chimborazo (ESPOCH) may be resolved. To accomplish this established goal, a Data Warehouse (DW) was implemented based on the require-ments of the UNAE to find reliable information through cleaning and data inte-gration techniques while respecting the Extraction, Transformation, and Load (ETL) process. In addition, several methods of DM were analyzed, culminating in the discovery of the pertinent information to ascertain the classification of stu-dents by areas of study, gender analysis, as well as to know the projection of the number of students who will commence university careers offered by ESPOCH in upcoming years.


CITT 2017 Organizing Committee
phone: (+593) 994420266 / 982910390 / 985292824

Visits counter

contador de visitas html