Cardiovascular disease (CVD) is the most common class of chronic and life-threatening diseases and, therefore, considered to be one of the main causes of mortality. The proposed new neural architecture based on the recent popularity of convolutional neural networks (CNN) was a solution for the development of automatic heart disease diagnosis systems using electrocardiogram (ECG) signals. More specifically, ECG signals were passed directly to a properly trained CNN network. The database consisted of more than 4000 ECG signal instances extracted from outpatient ECG examinations obtained from 47 subjects: 25 males and 22 females. The confusion matrix derived from the testing dataset indicated 99% accuracy for the “normal” class. For the “atrial premature beat” class, ECG segments were correctly classified 100% of the time. Finally, for the “premature ventricular contraction” class, ECG segments were correctly classified 96% of the time. In total, there was an average classification accuracy of 98.33%. The sensitivity (SNS) and the specificity (SPC) were, respectively, 98.33% and 98.35%. The new approach based on deep learning and, in particular, on a CNN network guaranteed excellent performance in automatic recognition and, therefore, prevention of cardiovascular diseases.

An accurate estimate of rainfall levels is fundamental in numerous application scenarios: weather forecasting, climate models, design of hydraulic structures, precision agriculture, etc. An accurate estimate becomes essential to be able to warn of the imminent occurrence of a calamitous event and reduce the risk to human beings. Unfortunately, to date, traditional techniques for estimating rainfall levels present numerous critical issues. The algorithm applies the Convolution Neural Network (CNN) directly to the audio signal, using 3 s sliding windows with an offset of only 100 milliseconds. Therefore, by using low cost and low power hardware, the proposed algorithm allows implementing critical high rainfall event alerting mechanisms with short response times and low estimation errors. More specifically, this paper proposes a new approach to rainfall estimation based on the classification of different acoustic timbres that rain produces at different intensities and on CNN. The results obtained on seven classes ranging from “No rain” to “Cloudburst” indicate an average accuracy of 75%, which rises to 93% if the misclassifications of the adjacent classes are not considered. Some application contexts concern smart cities for which the integration of an audio sensor inside the luminaire of a street lamp is foreseen, precision agriculture, as well as highway safety, by minimizing the risks of aquaplaning.

No abstract

The paper proposes an intelligent data sensing and geo-localization algorithm, based on an innovative mobile computing system that measures the power level of RF sources through a 2G/5G femtocell-UAV system. In natural disasters (mainly earthquakes and floods) the system can identify any missing persons under the rubble within a range of precision between 1 to 2 meters. In this paper, more specifically, the algorithm allows classifying the terminal even in the presence of obstacles that cause anisotropic propagation of radio signals, through a series of power measurements based on the Reference Signal Received Power (RSRP). An attenuation model that takes into account the different types of materials is introduced, and a method for optimizing the drone's flight path and duration is proposed. The performances, expressed in terms of accuracy in identifying the mobile terminal and in terms of position estimation average error, are evaluated according to the material's density and its attenuation. INDEX TERMS Femtocell/UAV system, geo-localization technique, 2G/5G radio technologies, reference signal received power (RSRP), mobile terminal classification/positioning algorithm, energy consumption.

The recent development of the IoT (Internet of Things), which has enabled new types of sensors that can be easily interconnected to the Internet, will also have a significant impact in the near future on the management of natural disasters (mainly earthquakes and floods) with the aim of improving effectiveness in research, identification, and recovery of missing persons, and therefore increasing the possibility of saving lives. In this paper, more specifically, an innovative technique is proposed for the search and identification of missing persons in natural disaster scenarios by employing a drone-femtocell system and devising an algorithm capable of locating any mobile terminal in a given monitoring area. In particular, through a series of power measurements based on the reference signal received power (RSRP), the algorithm allows for the classification of the terminal inside or outside the monitoring area and subsequently identify the position with an accuracy of about 1 m, even in the presence of obstacles that act in such a way as to make the propagation of the radio signal non-isotropic.