Search

Deep learning applications in medical imaging have been achieving promising results in the detection of diseases, among which clinical trials in terms of screening and diagnosis of patients with COVID-19 stand out. Computed Tomography (CT) images of the chest have been used by specialists for the diagnosis of COVID-19. However, due to the need of the moment and the possibility of using computational resources to help the medical team, it is observed in the literature several proposed works using supervised learning, however it lacks unsupervised methods for the screening and diagnosis of patients with COVID-19. In this work, the deep learning models Convolutional Neural Network (CNN) and Variational Autoencoders are used for feature extraction and later this information is used for binary and multiclass classification in unsupervised methods (k-means, Fuzzy C-Means and Self-Organizing Maps). For this purpose, a public database containing 4173 CT images (2168 CT slices from COVID-19, 758 slices from Healthy and 1247 slices from other lung diseases) was used. The results show that feature extraction via Variational Autoencoders has similar performance with state-of-the-art models in the literature for COVID-19, mainly for the binary classification with accuracies of 95.9%, 92.1% and 95.9% for k-means, Fuzzy C-Means and SOM, respectively, presenting competitive results in the literature. It also shows the importance of extracting features through convolutional networks to improve classification performance, resulting from the use of deep learning and its state of the art in the area of computer vision.

Light-field (LF) cameras, also known as plenoptic cameras, permit the recording of the 4D LF distribution of target scenes. However, many times, surface errors of a microlens array (MLA) are responsible for degradation in the images captured by a plenoptic camera. Additionally, the limited pixel count of the sensor can cause missing parallax information. The aforementioned issues are crucial for creating accurate maps for Intelligent Autonomous Transport System (IATS), because they cause loss of LF information, and need to be addressed. To tackle this problem, a learning-based framework by directly simulating the LF distribution is proposed. A high-dimensional convolution layer with densely sampled LFs in 4D space and considering a soft activation function based on ReLU segmentation correction is used to generate a superresolution (SR) LF image, improving the convergence rate in the deep learning network. Experimental results show that our proposed LF image reconstruction framework outperforms the existing state-of-the-art approaches; specifically, it is effective for learning the LF distribution and generating high-quality LF images. Different image quality assessment methods are used to evaluate the performance of the proposed framework, such as PSNR, SSIM, IWSSIM, FSIM, GFM, MDFM, and HDR-VDP. Additionally, the computational efficiency was evaluated in terms of number of parameters and FLOPs, and experimental results demonstrated that our proposed framework reached the highest performance in most of the datasets used.

The SARS-CoV-2 virus causes a respiratory disease in humans, known as COVID-19. The confirmatory diagnostic of this disease occurs through the real-time reverse transcription and polymerase chain reaction test (RT-qPCR). However, the period of obtaining the results limits the application of the mass test. Thus, chest X-ray computed tomography (CT) images are analyzed to help diagnose the disease. However, during an outbreak of a disease that causes respiratory problems, radiologists may be overwhelmed with analyzing medical images. In the literature, some studies used feature extraction techniques based on CNNs, with classification models to identify COVID-19 and non-COVID-19. This work compare the performance of applying pre-trained CNNs in conjunction with classification methods based on machine learning algorithms. The main objective is to analyze the impact of the features extracted by CNNs, in the construction of models to classify COVID-19 and non-COVID-19. A SARS-CoV-2 CT data-set is used in experimental tests. The CNNs implemented are visual geometry group (VGG-16 and VGG-19), inception V3 (IV3), and EfficientNet-B0 (EB0). The classification methods were k-nearest neighbor (KNN), support vector machine (SVM), and explainable deep neural networks (xDNN). In the experiments, the best results were obtained by the EfficientNet model used to extract data and the SVM with an RBF kernel. This approach achieved an average performance of 0.9856 in the precision macro, 0.9853 in the sensitivity macro, 0.9853 in the specificity macro, and 0.9853 in the F1 score macro.

Speech Emotion Recognition (SER) plays a vital role in human-computer interaction as an important branch of affective computing. Due to inconsistencies in the data and challenging signal extraction, in this paper, we propose a novel emotion recognition method based on the combination of Adaptive Neuro-Fuzzy Inference System (ANFIS) and Particle Swarm Optimization (PSO) with Word to Vector (Word2Vec) models. To begin, the inputs have been pre-processed, which comprise audio and text data. Second, the features were extracted using the Word2vec behind spectral and prosodic approaches. Finally, the features are selected using the Sequential Backward Floating Selection (SBFS) approach. In the end, the ANFIS-PSO model has been used to recognize speech emotion. A performance evaluation of the proposed algorithm is carried out on Sharif Emotional Speech Database (ShEMO). The experimental results show that the proposed algorithm has advantages in accuracy, reaching 0.873 and 0.752 in males and females, respectively, in comparison with the CNNs and SVM, MLP, RF models.

Nowadays, the high number of devices and applications connected to the Internet has generated a great amount of data being which makes privacy and protection a more challenging task. In addition, new technologies, such as the Internet of Things, incorporate many resource-constrained devices in the network. Reliable cryptography algorithms have to be employed to deal with this problem, which also needs to be efficiently implemented in small devices. There are several algorithms for this purpose, among them, neural cryptography. In this context, this work proposes the implementation of an artificial neural network architecture called tree parity machine (TPM) to perform the exchange of keys through the mutual learning of these networks. This method is not based on number theory, which makes it less computationally costly, and can be an alternative for embedded systems, which generally have several limitations in the processing capacity and resources used. In the area of embedded systems, FPGAs have gained more space, thanks to their reconfiguration capacity. Thus, different methods for implementing a TPM in FPGA were tested and analyzed, in order to optimize the following performance parameters, the response time, the maximum frequency of operation, and the consumed area of the FPGA considering logical elements, embedded multipliers, and registers. In addition, software implementation based on a multi-core CPU was used for comparison purposes. Experimental results demonstrated that the implementation of parallelism in FPGA for different blocks of the TPM weight matrix reached the best performance results. Thus, our proposal intends to develop an economic component in terms of resource consumption, however, maintaining the characteristic of high processing capacity. Therefore, the methodologies presented in this paper intends to be a useful reference to optimize future implementations in FPGA for cryptography applications.

Currently, with the increasing number of devices connected to the Internet, search for network vulnerabilities to attackers has increased, and protection systems have become indispensable. There are prevalent security attacks, such as the Distributed Denial of Service (DDoS), which have been causing significant damage to companies. However, through security attacks, it is possible to extract characteristics that identify the type of attack. Thus, it is essential to have fast and effective security identification models. In this work, a novel Intrusion Detection System (IDS) based on the Tree-CNN hierarchical algorithm with the Soft-Root-Sign (SRS) activation function is proposed. The model reduces the training time of the generated model for detecting DDoS, Infiltration, Brute Force, and Web attacks. For performance assessment, the model is implemented in a medium-sized company, analyzing the level of complexity of the proposed solution. Experimental results demonstrate that the proposed hierarchical model achieves a significant reduction in execution time, around 36%, and an average detection accuracy of 0.98 considering all the analyzed attacks. Therefore, the results of performance evaluation show that the proposed classifier based on Tree-CNN is of low complexity and requires less processing time and computational resources, outperforming other current IDS based on machine learning algorithms.

O processo de reconhecimento biométrico da íris é uma das tecnologias biométricas mais consistentes entre outras disponíveis atualmente. No entanto, sua eficiência e precisão podem ser afetadas por imagens de íris de baixa qualidade usadas como entrada para um sistema de reconhecimento, assim, o desempenho global é reduzido. Nesse contexto, este trabalho propõe um estudo de avaliação para determinar o impacto da qualidade da imagem da íris no desempenho do sistema biométrico da íris, utilizando as principais métricas apresentadas na norma ISO / IEC 29794-6: 2015. Os testes experimentais são realizados usando um banco de dados de imagens de íris e o software de reconhecimento biométrico OSIRIS, ambos amplamente aceitos e referenciados nas últimas pesquisas. Os resultados experimentais mostram os valores de intervalo de cada métrica de qualidade e o número de imagens que atingem os valores mínimos necessários. O desempenho do sistema biométrico é avaliado pelos parâmetros True-Match (TM) e False Non-Match (FNM); assim, foi possível identificar que quanto maior o nível de qualidade da imagem, menor o valor de FNM; portanto, o desempenho do sistema é aprimorado.

Many factors can affect the users’ quality of experience (QoE) in speech communication services. The impairment factors appear due to physical phenomena that occur in the transmission channel of wireless and wired networks. The monitoring of users’ QoE is important for service providers. In this context, a non-intrusive speech quality classifier based on the Tree Convolutional Neural Network (Tree-CNN) is proposed. The Tree-CNN is an adaptive network structure composed of hierarchical CNNs models, and its main advantage is to decrease the training time that is very relevant on speech quality assessment methods. In the training phase of the proposed classifier model, impaired speech signals caused by wired and wireless network degradation are used as input. Also, in the network scenario, different modulation schemes and channel degradation intensities, such as packet loss rate, signal-to-noise ratio, and maximum Doppler shift frequencies are implemented. Experimental results demonstrated that the proposed model achieves significant reduction of training time, reaching 25% of reduction in relation to another implementation based on DRBM. The accuracy reached by the Tree-CNN model is almost 95% for each quality class. Performance assessment results show that the proposed classifier based on the Tree-CNN overcomes both the current standardized algorithm described in ITU-T Rec. P.563 and the speech quality assessment method called ViSQOL.

People use Online Social Networks (OSNs) to express their opinions and feelings about many topics. Depending on the nature of an event and its dissemination rate in OSNs, and considering specific regions, the users’ behavior can drastically change over a specific period of time. In this context, this work aims to propose an event detection system at the early stages of an event based on changes in the users’ behavior in an OSN. This system can detect an event of any subject, and thus, it can be used for different purposes. The proposed event detection system is composed of the following main modules: (1) determination of the user’s location, (2) message extraction from an OSN, (3) topic identification using natural language processing (NLP) based on the Deep Belief Network (DBN), (4) the user behavior change analyzer in the OSN, and (5) affective analysis for emotion identification based on a tree-convolutional neural network (tree-CNN). In the case of public health, the early event detection is very relevant for the population and the authorities in order to be able take corrective actions. Hence, the new coronavirus disease (COVID-19) is used as a case study in this work. For performance validation, the modules related to the topic identification and affective analysis were compared with other similar solutions or implemented with other machine learning algorithms. In the performance assessment, the proposed event detection system achieved an accuracy higher than 0.90, while other similar methods reached accuracy values less than 0.74. Additionally, our proposed system was able to detect an event almost three days earlier than the other methods. Furthermore, the information provided by the system permits to understand the predominant characteristics of an event, such as keywords and emotion type of messages.

Voice quality assessment of phone calls is a relevant task for mobile service providers. In this context, the main objective of this research is to provide a model that improves the performance of ITU-T Rec. P.563. To accomplish this objective, the proposed model considers two aspects, better response in lossy network and adequate treatment of silences segments into the audio signal. Thus, a function is determined to suppress silences in the speech signal according to the packet loss rate value. Furthermore, the proposed model is implemented on both a voice quality server and a mobile device. Experimental results show that P.563 algorithm performance was really improved by the proposed model, approximating its results to those given by P.862 algorithm, reaching a Pearson correlation coefficient of 0.9957 and a root mean square error of 0.2983. Moreover, subjective test results demonstrated that the proposed model results overcome those obtained by the P.563 algorithm.

Nowadays, one of the most outstanding new urban transport model is the ridesharing service, in which two or more users share a ride. This transport model reduces costs and the number of circulating vehicles, improving user mobility. In the ridesharing service, the users’ quality is a tangible novel evaluation parameter. Consequently, this study treats the use of quality of experience (QoE) in the ridesharing service context, proposing a recommendation system (RS) for ridesharing services (RsRS), which considers user profile information extracted from online social networks (OSN) and user preferences. Thus, the main objective of the proposed RsRS is to improve users’ QoE. To this end, the users’ profile for the ridesharing service is built based on OSN data, which includes group of users with similar characteristics in the same trip, thus avoiding users with opposite preferences. First, subjective tests are carried out to obtain information on users’ preferences and the results are analyzed via machine learning algorithms to determine the various user profiles. The experimental results demonstrate that the random forest algorithm has the best performance, considering OSN data and explicit preferences saved in the proposed solution and only OSN data, for average F-measures of 0.92 and 0.91, respectively. Additionally, a ranking containing a list of recommended users to share a ride is determined using a similarity function, and the results demonstrate that 94.2% of assessors agree with the proposed recommendations. Furthermore, the RsRS has a modular configuration and its integration with a real ridesharing service providers is also discussed.

Made available in DSpace on 2019-10-04T15:23:58Z (GMT). No. of bitstreams: 0 Previous issue date: 2019-04-01 Online social networks provide relevant information on users' opinion about different themes. Thus, applications, such as monitoring and recommendation systems (RS) can collect and analyze this data. This paper presents a knowledge-based recommendation system (KBRS), which includes an emotional health monitoring system to detect users with potential psychological disturbances, specifically, depression and stress. Depending on the monitoring results, the KBRS, based on ontologies and sentiment analysis, is activated to send happy, calm, relaxing, or motivational messages to users with psychological disturbances. Also, the solution includes a mechanism to send warning messages to authorized persons, in case a depression disturbance is detected by the monitoring system. The detection of sentences with depressive and stressful content is performed through a convolutional neural network and a bidirectional long short-term memory recurrent neural networks (RNN); the proposed method reached an accuracy of 0.89 and 0.90 to detect depressed and stressed users, respectively. Experimental results show that the proposed KBRS reached a rating of 94% of very satisfied users, as opposed to 69% reached by a RS without the use of neither a sentiment metric nor ontologies. Additionally, subjective test results demonstrated that the proposed solution consumes low memory, processing, and energy from current mobile electronic devices. Univ Fed Lavras, BR-37200000 Lavras, MG, Brazil Univ Estadual Paulista, Biosci Inst Rio Claro, BR-13506900 Sao Paulo, Brazil Univ Sao Paulo, Polytech Sch, BR-05508010 Sao Paulo, Brazil Univ Estadual Paulista, Biosci Inst Rio Claro, BR-13506900 Sao Paulo, Brazil

Inter-satellite optical wireless communication (Is-OWC) links can become promising solutions to realize the next-generation high-speed communication services. The operation of Global Navigation Satellite Systems can be improved with the use of Is-OWC links through ranging and communication services. However, the key challenge in Inter-satellite link (ISL) is its effective range which is limited due to pointing errors. In this work, we propose to develop a high-capacity and long-reach Is-OWC link by incorporating hybrid mode division multiplexing (MDM) and wavelength division multiplexing (WDM) schemes to transmit ten independent channels over 40000kms Is-OWC link. Each channel is capable of carrying 400Gbps data which is encoded by the dual polarization quadrature phase shift key technique with required signal to noise ratio (SNR) and received power. The proposed Is-OWC link satisfies the enhanced communication within Geostationary Earth Orbit (GEO) and Low Earth Orbit (LEO) satellites. The proposed Is-OWC is further evaluated under the impact of space turbulences, particularly transmitter and receiver pointing errors. The result reported that the proposed Is-OWC link can transmit 4Tbps data over 16000kms with the transmitter pointing error of 2μrad and receiver pointing error of 1μrad.

Light field (LF) imaging has multi-view properties that help to create many applications that include auto-refocusing, depth estimation and 3D reconstruction of images, which are required particularly for intelligent transportation systems (ITSs). However, cameras can present a limited angular resolution, becoming a bottleneck in vision applications. Thus, there is a challenge to incorporate angular data due to disparities in the LF images. In recent years, different machine learning algorithms have been applied to both image processing and ITS research areas for different purposes. In this work, a Lightweight Deformable Deep Learning Framework is implemented, in which the problem of disparity into LF images is treated. To this end, an angular alignment module and a soft activation function into the Convolutional Neural Network (CNN) are implemented. For performance assessment, the proposed solution is compared with recent state-of-the-art methods using different LF datasets, each one with specific characteristics. Experimental results demonstrated that the proposed solution achieved a better performance than the other methods. The image quality results obtained outperform state-of-the-art LF image reconstruction methods. Furthermore, our model presents a lower computational complexity, decreasing the execution time.

Telecommunication service providers have to guarantee acceptable speech quality during a phone call to avoid a negative impact on the users' quality of experience. Currently, there are different speech quality assessment methods. ITU-T Recommendation G.107 describes the E-model algorithm, which is a computational model developed for network planning purposes focused on narrowband (NB) networks. Later, ITU-T Recommendations G.107.1 and G.107.2 were developed for wideband (WB) and fullband (FB) networks. These algorithms use different impairment factors, each one related to different speech communication steps. However, the NB, WB, and FB E-model algorithms do not consider wireless techniques used in these networks, such as Multiple-Input-Multiple-Output (MIMO) systems, which are used to improve the communication system robustness in the presence of different types of wireless channel degradation. In this context, the main objective of this study is to propose a general methodology to incorporate wireless network parameters into the NB and WB E-model algorithms. To accomplish this goal, MIMO and wireless channel parameters are incorporated into the E-model algorithms, specifically into the $I_{e,eff}$ and $I_{e,eff,WB}$ impairment factors. For performance validation, subjective tests were carried out, and the proposed methodology reached a Pearson correlation coefficient (PCC) and a root mean square error (RMSE) of $0.9732$ and $0.2351$, respectively. It is noteworthy that our proposed methodology does not affect the rest of the E-model input parameters, and it intends to be useful for wireless network planning in speech communication services., 18 pages

A quality monitoring system for telecommunication services is relevant for network operators because it can help to improve users’ quality-of-experience (QoE). In this context, this article proposes a quality monitoring system, named Q-Meter, whose main objective is to improve subscriber complaint detection about telecommunication services using online-social-networks (OSNs). The complaint is detected by sentiment analysis performed by a deep learning algorithm, and the subscriber’s geographical location is extracted to evaluate the signal strength. The regions in which users posted a complaint in OSN are analyzed using a freeware application, which uses the radio base station (RBS) information provided by an open database. Experimental results demonstrated that sentiment analysis based on a convolutional neural network (CNN) and a bidirectional long short-term memory (BLSTM)-recurrent neural network (RNN) with the soft-root-sign (SRS) activation function presented a precision of 97% for weak signal topic classification. Additionally, the results showed that 78.3% of the total number of complaints are related to weak coverage, and 92% of these regions were proved that have coverage problems considering a specific cellular operator. Moreover, a Q-Meter is low cost and easy to integrate into current and next-generation cellular networks, and it will be useful in sensing and monitoring tasks.

The routing algorithm is one of the main factors that directly impact on network performance. However, conventional routing algorithms do not consider the network data history, for instances, overloaded paths or equipment faults. It is expected that routing algorithms based on machine learning present advantages using that network data. Nevertheless, in a routing algorithm based on reinforcement learning (RL) technique, additional control message headers could be required. In this context, this research presents an enhanced routing protocol based on RL, named e-RLRP, in which the overhead is reduced. Specifically, a dynamic adjustment in the Hello message interval is implemented to compensate the overhead generated by the use of RL. Different network scenarios with variable number of nodes, routes, traffic flows and degree of mobility are implemented, in which network parameters, such as packet loss, delay, throughput and overhead are obtained. Additionally, a Voice-over-IP (VoIP) communication scenario is implemented, in which the E-model algorithm is used to predict the communication quality. For performance comparison, the OLSR, BATMAN and RLRP protocols are used. Experimental results show that the e-RLRP reduces network overhead compared to RLRP, and overcomes in most cases all of these protocols, considering both network parameters and VoIP quality.

The present work focuses on the modeling of the thermal system composed of several modules, in which the general model of the system is built, which includes all its components in operation. The environment collected data in real time of the growth and decrease (temperature) curve of a soldering iron tool, under aspects of PWM variation - Pulse Width Modulation - of the signal. It is used as a representative model of device in industrial welding, in which it has similarities in key aspects of behavior. Through analysis and software assistance, the function of transferring the general plan was obtained, which showed an acceptable level of correspondence (86.04%) in non-linear systems. Based on the model discovered, a PItype controller - Proportional Integral - was designed, paying attention to the need for a low-level overshoot signal (6.08%) in order to obtain a response to the rapid stabilization step, rare in thermal systems. The PI was developed in order to generate an output in DC format (Duty Cycle format), since the raw data of the system were collected by varying this indicator at regular intervals of time. After the tests carried out in simulation, results were obtained that confirm that the developed control system is applicable and responds appropriately, being able to be optimized for applications.

In this paper, we propose a global digital platform to avoid and combat epidemics by providing relevant real-time information to support selective lockdowns. It leverages the pervasiveness of wireless connectivity while being trustworthy and secure. The proposed system is conceptualized to be decentralized yet federated, based on ubiquitous public systems and active citizen participation. Its foundations lie on the principle of informational self-determination. We argue that only in this way it can become a trustworthy and legitimate public good infrastructure for citizens by balancing the asymmetry of the different hierarchical levels within the federated organization while providing highly effective detection and guiding mitigation measures towards graceful lockdown of the society. To exemplify the proposed system, we choose the remote patient monitoring as use case. In which, the integration of distributed ledger technologies with narrowband IoT technology is evaluated considering different number of endorsed peers. An experimental proof of concept setup is used to evaluate the performance of this integration, in which the end-to-end latency is slightly increased when a new endorsed element is added. However, the system reliability, privacy, and interoperability are guaranteed. In this sense, we expect active participation of empowered citizens to supplement the more usual top-down management of epidemics., 9 pages, 3 figures, Accepted for Publication

Communication service providers use specialized solutions to evaluate the quality of their services. Also, different mechanisms that increase network robustness are incorporated in current communication systems. One of the most accepted techniques to improve transmission performance is the MIMO system. In communication services, voice quality is important to determine the user's quality of experience. Nowadays, different speech quality assessment methods are used, one of them is the parametric method that is used for network planning purposes. The ITU-T Rec. G107.1 is the most accepted model for wide-band communication systems. However, it does not consider the degradations occurring in a wireless network nor the quality improvement, caused by the MIMO systems. Thus, we propose a speech quality model, based on wireless parameters, such as signal-to-noise ratio, Doppler shift, MIMO configurations, and different modulation schemes. Also, real speech signals encoded by 3 operation modes of the AMR-WB codec are used in test scenarios. The resulting speech samples were evaluated by the algorithm described in ITU-T Rec. P.862.2, which scores are used as a reference. With these results, a wireless function, named IW-M that relates the wireless network parameters with speech quality is proposed and inserted into the wide-band E-model algorithm. It is worth noting that the main novelty of the proposed IW-M is the consideration of the quality improvement incorporated by MIMO systems with different antenna array configurations. The performance validation results demonstrated that the inclusion of IW-M values into the R global score determined a reliable model, reaching a Pearson correlation coefficient and a normalized RMSE of 0.976 and 0.144, respectively.

Nowadays, the voice over Internet protocol (VoIP) communication service is widely adopted, and it counts with many users across the world. However, the users’ quality of experience is not guaranteed because the voice signal quality can be affected by several degradations that happen in the network infrastructure. Thus, it is relevant to have a global speech quality assessment method that considers both wired and wireless networks to provide reliable results. In this paper, several network scenarios that consider different packet loss rates (PLRs) and wireless channel models are implemented in which the impaired signals are evaluated using the algorithm described in ITU-T Recommendation P.862. Preliminary results showed a relationship between both fading and PLR parameters and the global speech quality index. However, the P.862 algorithm is not viable in real VoIP scenarios. The ITU-T Recommendation P.563 describes a non-intrusive speech quality assessment method; nevertheless, its results are not confident. In this context, the main objective of this paper is to propose a non-intrusive speech quality classification model based on a deep belief network (DBN) that considers the wired and wireless impairments on the speech signal. Experimental results demonstrated a high correlation between the proposed model based on the DBN and P.862 algorithm, reaching a ${F}$ -measure of 97.01%. For validation, the non-intrusive P.563 algorithm is used; the proposed model and P.563 reached an average accuracy of 96.14% and 72.12%, respectively. Furthermore, subjective tests were carried out, and the proposed DBN model reached an accuracy of 94%.

Nowadays, there are several telephone services based on IP networks. However, the networks can present many disturbances, such as packet loss rate (PLR), which is one of the most impairing network factors. An impaired speech communication affects the users’ quality of experience; hence, the assessment of speech quality is relevant to the telephone operators. Therefore, the determination of a methodology to predict a speech quality with a higher accuracy in telephone services is relevant. In this context, this letter introduces a novel nonintrusive speech quality classifier (SQC) model based on deep belief networks (DBN), in which the support vector machine with radial basis function kernel is the classifier applied in DBN, in order to identify four speech quality classes. A speech database was built, based on unimpaired speech files of public databases, in which different PLR models and values are applied, and a standardized intrusive method is used to calculate the index quality of each file. Results show that SQC largely overcomes the results obtained by ITU-T Recommendation P.563. Also, subjective tests are performed to validate the SQC performance, and it reached an accuracy of 95% on speech quality classification. Furthermore, a solution architecture is introduced, demonstrating the usefulness and flexibility of the proposed SQC.

Current implementations of 5G networks consider higher frequency range of operation than previous telecommu nication networks, and it is possible to offer higher data rates for different applications. On the other hand, atmospheric phe nomena could have a more negative impact on the transmission quality. Thus, the study of the transmitted signal quality at high frequencies is relevant to guaranty the user s quality of experience. In this research, the recommendations ITU-R P.838- 3 and ITU-R P.676-11 are implemented in a network scenario, which are methodologies to estimate the signal degradations originated by rainfall and atmospheric gases, respectively. Thus, speech signals are encoded by the AMR-WB codec, transmitted and the perceptual speech quality is evaluated using the algorithm described in ITU-T Rec. P.863, mostly known as POLQA. The novelty of this work is to propose a non-intrusive speech quality classifier that considers atmospheric phenomena. This classifier is based on Deep Belief Networks (DBN) that uses Support Vector Machine (SVM) with radial basis function kernel (RBF-SVM) as classifier, to identify five predefined speech quality classes. Experimental Results show that the proposed speech quality classifier reached an accuracy between 92% and 95% for each quality class overcoming the results obtained by the sole non- intrusive standard described in ITU-T Recommendation P.563. Furthermore, subjective tests are carried out to validate the proposed classifier performance, and it reached an accuracy of 94.8%.

Modern communication systems, such as 5G networks promise to meet the crescent demand of users for high data rates in order to support new technologies. However, the frequency range of operation of these networks deserves special attention, since the high frequencies are more susceptible to interferences caused by atmospheric phenomena. The recommendations ITU-R P.838-3 and ITU-R P.676-11 describe methodologies to estimate signal degradations caused by phenomena related to rain and atmospheric gases, respectively. In this work, the impact of such phenomena on speech signal quality is investigated. The perceptual speech quality is evaluated using the algorithm described in ITU-T Rec. P.862. The experimental results show that the higher the frequency of operation, the higher the level of signal degradation for the same atmospheric conditions. Based on these results a parametric speech quality assessment model is proposed that uses an artificial neural network. Performance evaluation results demonstrated a high correlation between the proposed model and subjective test results, reaching an PCC and an RMSE of 0.9869 and 0.4773, respectively. Hence, the proposed model intends to be useful for network planning tasks.

Currently, an access point (AP) is usually selected based on the signal strength parameter. However, the signal strength is not a guarantee of a good quality of service (QoS). Machine learning algorithms are used to automatically learn and improve some tasks and based on a network device characteristics is possible to select the most important input for a better network coverage. Thus, in this paper is implemented a Resource Allocation service for wireless networks based on machine learning algorithms. In this research, the Random Forest algorithm was implemented to automatically determine the AP selection strategy (SS). The results of the RF algorithm applied to heterogeneous network technologies showed an improvement of the channel condition, in relation to the throughput. In the validation tests phase, the experimental results demonstrated that our proposed AP SS based on Random Forest algorithm outperforms an existing AP SS based on signal strength.

Daily, people have access to various multimedia content, that can change their emotional state and it can cause an increase in their heart rate. These emotional changes are troubling in the elderly people. Thus, this work aims to classify emotions by machine learning and monitor the emotions of the elderly through the measurement of heart rate, using portable sensors of low cost and easily to use. For this, a framework is implemented, which presents to the user messages of attention and danger according to emotion and heart rate when video contents are presented to the elderly people. Also, the audio transcription of the videos are performed and the sentiment analysis is performed. Thus, when detecting a sudden increase in heart rate and the video presents a negative sentiment polarity, then the proposed solution presents a video of calm or happy content, according to the user's pre-registered preference. The experimental results showed that the classification system performed by a deep learning model obtained values of 97% of accuracy for classifying the disgust emotion and the heart rate level decreased with the proposed framework. Additionally, 93.75% of the users considered the system efficient.

This work presents the development of a wireless Access Point (AP) placement software that uses microwave signal propagation models. We use the Simulated Annealing metaheuristic, to improve the signal coverage according to the physical features of the architectural floor plan. In the experiments, a Graphics Processing Unit (GPU) with CUDA is used to parallelize the simulations, speeding up the runtime required. It is worth noting that our software makes possible to test the provisions of APs without the operational cost of having to move them physically. It proposes a spatial arrangement of APs that provides a greater coverage with usable signal intensity within the simulated environment. Using a limited amount of APs, the test case results showed that it is possible to obtain a significant improvement in the coverage of the Wi-Fi signal by adjusting the APs positions to the metaheuristic proposed solution. In addition, our software is capable of performing simulations with an increasing number of APs if the current quantity is not enough to achieve better quality and signal coverage.

In communication services, speech quality plays an important role to achieve user expectations. Nowadays, there are different objective methods to estimate speech quality. Parametric models consider different factors, such as network parameters, acoustic characteristics, communication equipment, among others. The most representative parametric models for telephone service are described in ITU-T Rec. G.107 and G.107.1, mostly known as E-model and WB E-model, respectively. However, they do not consider wireless network parameters as inputs. In this context, this research proposes a speech quality parametric model (SQPM) based on artificial neural networks that considers both wireless network degradation characteristics and the techniques used to improve the transmission quality. For this purpose, a network simulator was built, in which two forward error correction (FEC) codes and four different antenna configurations in a multiple-input-multiple-output (MIMO) system are implemented. To validate the results obtained by the simulator, the ITU-T Rec. P.863 and the WB E-model are used. Experimental results show how different wireless network configurations impact on speech quality. Performance evaluation results demonstrated a high correlation between the proposed SQPM and ITU-T Rec. P.863 results, reaching an PCC and an RMSE of 0.9901 and 0.1492, respectively. Therefore, our proposal intends to be useful for wireless network planning tasks.

In current communication systems, forward Error Correction (FEC) codes are used to decrease the information losses in the transmission channel to improve the signal quality in the reception point. The speech quality is affected by many factors being packet losses one of the most important. Currently, there are different speech quality assessment methods; for planning purposes, E-model algorithm is the most representative. However, it does not consider wireless network characteristics and techniques. Based on this fact, the main contribution of this work is to adapt the wide-band (WB) E-model algorithm to be able to evaluate the impact of FEC codes on speech quality. For this purpose, a function named G FECx(Ms,SNR) is proposed that quantifies the quality gain reached by FEC codes at different wireless channel conditions. This function is inserted into I e,eff,WB impairment factor of the WB E-model algorithm. Experimental results demonstrated that the proposed solution gets a high correlation with results obtained by ITU-T Rec. P.863, reaching an average PCC and RMSE of 0.990 and 2.629, respectively. Thus, the methodology introduced in this research makes it possible to estimate the speech quality when FEC codes are implemented in a communication system.

O amortecimento é uma propriedade dos materiais que pode ocorrer em aspecto de escala micro e macroscópica. Há um crescimento no conceito da análise de marcadores de trincas internas estruturais que causam danos em materiais diversos, devido sua ampla aplicabilidade na resolução de problemas. No processo de amortecimento ocorre a perda de energia do material quando este entra em contato físico com outro material, no qual o fenômeno pode ser observado por meio de equipamentos modernos de alta sensibilidade sob ambiente controlado. Há poucos institutos de pesquisa com capacidade de análises em ensaios de amortecimento precisos com corpos de prova. O uso de dispositivo acelerômetro contorna a problemática do alto custo laboratorial envolvidos e permite maior acessíbilidade. Objetiva-se apresentar uma metodologia de análise de parâmetros associados ao fenômeno de amortecimento, fazendo-se uso de algotítmos para acesso aberto e que permita sua replicação em modelos físicos. O trabalho apresentou efciência em determinar a frequência e parâmetros associados de sinais de corpos de prova.

Nos serviços de comunicação, a qualidade e integridade do sinal de voz no receptor é um fator relevante que as operadoras de telefonia devem considerar. Neste artigo, estudamos o impacto da degradação que ocorre no canal de transmissão no sinal de voz transmitido. Para tanto, são utilizados parâmetros-chave, como a pontuação média de opiniões (MOS) relacionada ao sinal de voz e a taxa de erro de bit (BER), a fim de avaliar qualitativa e quantitativamente o sinal de fala em cada cenário. É usado um cenário de simulação de modulação, que considera os canais de desvanecimento AWGN, Rician e Rayleigh, e duas modulações: BPSK e QAM. Esses canais têm diferentes parâmetros de configuração. Os resultados experimentais mostram que existem certos valores de índice de qualidade MOS para cada tipo de canal com uma configuração específica, mostrando que para cada cenário de transmissão há a necessidade de adaptações na combinação modulação-canal-ganho para obter melhor desempenho.

Social networks have a large amount of data available, but often, people do not provide some of their personal data, such as age, gender, and other demographics. Although the sentiment analysis uses such data to develop useful applications in people’s daily lives, there are still failures in this type of analysis, either by the restricted number of words contained in the word dictionaries or because they do not consider the most diverse parameters that can influence the sentiments in a sentence; thus, more reliable results can be obtained, if the users profile information and their writing characteristics are considered. This research suggests that one of the most relevant parameter contained in the user profile is the age group, showing that there are typical behaviors among users of the same age group, specifically, when these users write about the same topic. A detailed analysis with 7000 sentences was performed to determine which characteristics are relevant, such as, the use of punctuation, number of characters, media sharing, topics, among others; and which ones can be disregarded for the age groups classification. Different learning machine algorithms are tested for the classification of the teenager and adult age group, and the deep convolutional neural network had the best performance, reaching a precision of 0.95 in the validation tests. Furthermore, in order to validate the usefulness of the proposed model for classifying age groups, it is implemented into the enhanced sentiment metric (eSM). In the performance validation, subjective tests are performed and the eSM with the proposed model reached a root mean square error and a Pearson correlation coefficient of 0.25 and 0.94, respectively, outperforming the eSM metric, when the age group information is not available.

Currently, dynamic adaptive streaming over HTTP (DASH) standard is used to change the video resolution according to the network capacity of the end user’s. However, if the video resolution changes frequently, the attention of the users can be affected decreasing their quality-of-experience (QoE). In this context, this paper proposes a no-reference video quality metric that takes three impairment factors into account, initial buffering delay, temporal interruptions or pauses, and video resolution changes during a video transmission. Also, the temporal locations of pauses and resolution changes are considered. In order to perform this task, extensive subjective tests were conducted. Experimental results show that the evaluators’ QoE is highly correlated with the frequency of pauses, video resolution changes and initial delay. Based on these results a video streaming quality metric for DASH services (VsQM $_{\text {DASH}}$ ) is formulated and implemented on PC and mobile hand-held devices. Finally, the performance of VsQM $_{\text {DASH}}$ is validated with additional subjective tests reaching a person correlation coefficient of 0.99. Furthermore, experimental results demonstrated that the proposed metric has low complexity because it is based on application level parameters and its energy consumption is negligible in relation to the energy consumed by the video player; therefore, can be easily implemented in consumer electronic devices.

Minimizing human intervention in engines, such as traffic lights, through automatic applications and sensors has been the focus of many studies. Thus, Deep Learning (DL) algorithms have been studied for traffic signs and vehicle identification in an urban traffic context. However, there is a lack of priority vehicle classification algorithms with high accuracy, fast processing, and a lightweight solution. For filling those gaps, a vehicle detection system is proposed, which is integrated with an intelligent traffic light. Thus, this work proposes (1) a novel vehicle detection model named Priority Vehicle Image Detection Network (PVIDNet), based on YOLOV3, (2) a lightweight design strategy for the PVIDNet model using an activation function to decrease the execution time of the proposed model, (3) a traffic control algorithm based on the Brazilian Traffic Code, and (4) a database containing Brazilian vehicle images. The effectiveness of the proposed solutions were evaluated using the Simulation of Urban MObility (SUMO) tool. Results show that PVIDNet reached an accuracy higher than 0.95, and the waiting time of priority vehicles was reduced by up to 50%, demonstrating the effectiveness of the proposed solution.

Nowadays, shared mobility service is a trend in many countries. It tends to grow even more because of its low cost, the mitigation of both traffic and pollution, and due to the spreading of several shared-use mobility applications on mobile devices. As seen in other services, the success is based on the satisfaction level attained by users. Hence, if a ride is shared between people with similar preferences, users will feel more comfortable and safer. However, finding users with similar preferences is still a challenge in shared-use mobility services. In this context, this research shows that using some basic user information, such as gender, age, and relationship, extracted from Online Social Networks(OSN), and also some preferences, it is possible to determine if the user wants to share a vehicle with people with specific characteristics. Thus, as contribution, it is possible to classify users of similar preferences, automatically, to improve their ride experience. The classification was performed through machine learning algorithms, in which a Discriminative Restricted Boltzmann Machine(DRBM) algorithm reaches a correct classified instance of 94.5% and F-Measure of 0.93 for the option of sharing a ride with a person with similar hobby. Then, a Recommendation System(RS) is proposed, which efficiency is compared with a basic RS; they reached a Pearson Correlation Coefficient of 0.96 and 0.79, respectively; highlighting the importance of considering user preferences. Also, it is important to note that this study can be extended for other sharing services.