This work introduces a new approach for automatic identification of crickets, katydids and cicadas analyzing their acoustic signals. We propose the building of a tool to identify this biodiversity. The study proposes a sound parameterization technique designed specifically for identification and classification of acoustic signals of insects using Mel Frequency Cepstral Coefficients (MFCC) and Linear Frequency Cepstral Coefficients (LFCC). These two sets of coefficients are evaluated individually as has been done in previous studies and have been compared with the fusion proposed in this work, showing an outstanding increase in identification and classification at species level reaching a success rate of 98.07% on 343 insect species.

Fish as well as birds, mammals, insects and other animals are capable of emitting sounds for diverse purposes, which can be recorded through microphone sensors. Although fish vocalizations have been known for a long time, they have been poorly studied and applied in their taxonomic classification. This work presents a novel approach for automatic remote acoustic identification of fish through their acoustic signals by applying pattern recognition techniques. The sound signals are preprocessed and automatically segmented to extract each call from the background noise. Then, the calls are parameterized using Linear and Mel Frequency Cepstral Coefficients (LFCC and MFCC), Shannon Entropy (SE) and Syllable Length (SL), yielding useful information for the classification phase. In our experiments, 102 different fish species have been successfully identified with three widely used machine learning algorithms: K-Nearest Neighbors (KNN), Random Forest (RF) and Support Vector Machine (SVM). Experimental results show an average classification accuracy of 95.24%, 93.56% and 95.58%, respectively.

In recent years, there has been a remarkable advance in monitoring technologies in many environments, be they urban or rural. These technologies, included in the Internet of Things (IoT) domain, allow remote control and acquisition of data from sensors for their subsequence analysis. All these systems are based on the interaction between sensors and actuators. To achieve this goal, it is necessary to provide a very high level of connectivity between the devices, especially as far as wireless systems are concerned. In this sense, there is a great variety of standards in the market of communication networks oriented to this end. One of the biggest challenges today is to allow inter-operability between these different technologies in order to homogenize this field. In addition to this, it is intended to introduce new communication techniques that can provide certain additional advantages to those already existing. The main idea is the creation of a cellular network where radiofrequency and optical technologies coexist, and whose link with the rest of the world is through long-range and low-consumption wireless technologies. The center of each cell, that is the lighting system, can be powered using solar panels, as can the existing systems in the market. The objective is that these panels are capable of providing the necessary energy to the rest of the necessary systems.

Indoor position estimation has become an attractive research topic due to growing interest in location-aware services. Nevertheless, satisfying solutions have not been found with the considerations of both accuracy and system complexity. From the perspective of lightweight mobile devices, they are extremely important characteristics, because both the processor power and energy availability are limited. Hence, an indoor localization system with high computational complexity can cause complete battery drain within a few hours. In our research, we use a data mining technique named boosting to develop a localization system based on multiple weighted decision trees to predict the device location, since it has high accuracy and low computational complexity. The localization system is built using a dataset from sensor fusion, which combines the strength of radio signals from different wireless local area network access points and device orientation information from a digital compass built-in mobile device, so that extra sensors are unnecessary. Experimental results indicate that the proposed system leads to substantial improvements on computational complexity over the widely-used traditional fingerprinting methods, and it has a better accuracy than they have.

Indoor localization estimation has become an attractive research topic due to growing interest in location-aware services. Many research works have proposed solving this problem by using wireless communication systems based on radiofrequency. Nevertheless, those approaches usually deliver an accuracy of up to two metres, since they are hindered by multipath propagation. On the other hand, in the last few years, the increasing use of light-emitting diodes in illumination systems has provided the emergence of Visible Light Communication technologies, in which data communication is performed by transmitting through the visible band of the electromagnetic spectrum. This brings a brand new approach to high accuracy indoor positioning because this kind of network is not affected by electromagnetic interferences and the received optical power is more stable than radio signals. Our research focus on to propose a fingerprinting indoor positioning estimation system based on neural networks to predict the device position in a 3D environment. Neural networks are an effective classification and predictive method. The localization system is built using a dataset of received signal strength coming from a grid of different points. From the these values, the position in Cartesian coordinates $(x,y,z)$ is estimated. The use of three neural networks is proposed in this work, where each network is responsible for estimating the position by each axis. Experimental results indicate that the proposed system leads to substantial improvements to accuracy over the widely-used traditional fingerprinting methods, yielding an accuracy above 99% and an average error distance of 0.4 mm.