Automated document classification through machine learning is crucial for handling extensive modern datasets, as manual analysis becomes impractical due to time and cost constraints. Real-world datasets often grapple with issues like class imbalance and high dimensionality, challenging classification algorithms. Our multi-class classification approach encompasses pre-processing, data balancing, feature engineering, and dimensionality reduction. We conducted experiments on seven datasets, including benchmarks and private data, leveraging natural language processing (NLP) techniques like lemmatization and stemming for corpus preparation. To mitigate feature dimensionality, we applied principal component analysis (PCA) and employed oversampling to tackle data imbalance. Finally, we utilized multilayer perceptron neural networks (MLP), support vector machines (SVM), and random forest (RF) models for clas-sification. Our framework achieved high accuracy, ranging from 77 % to 99 %, contingent on the dataset, while also demonstrating strong performance across additional metrics like precision, F1-score, and recall.

Drivers' behavior in traffic is a determining factor for the rate of accidents on roads and highways.This paper presents the design of an intelligent IoT system capable of inferring and warning about road traffic risks and danger zones, based on data obtained from the vehicles and their drivers mobile phones, thus helping to avoid accidents and seeking to preserve the lives of the passengers.The proposed approach is to collect vehicle telemetry data and mobile phone sensors data through an IoT network and then to analyze the driver's behavior while driving, along with data from the environment.The results of the inference serve to alert drivers about incidents in their trajectory as well as to provide feedback on how they are driving.The proposal is validated using a developed prototype to test its data collection and inference features in a small scale experiment.

In this paper, we address the problem of joint downlink (DL) and uplink (UL) channel estimation for millimeter wave (mmWave) multiple-input multiple-output (MIMO) systems. Assuming a closed-loop and multifrequency-based channel training framework in which pilot signals received by multiple antenna mobile stations (MSs) are coded and spread in the frequency domain via multiple adjacent subcarriers, we propose two tensor-based semiblind receivers by capitalizing on the multilinear structure and sparse feature of the received signal at the BS equipped with a hybrid analog-digital beamforming (HB) architecture. As a first processing stage, the joint estimation of the compressed DL and UL channel matrices can be obtained in an iterative way by means of an alternating least squares (ALS) algorithm that capitalizes on a parallel factors model for the received signals. Alternatively, for more restricted scenarios, a closed-form solution is also proposed. From the estimated effective channel matrices, the users’ channel parameters such as angles of departure (AoD), angles of arrival (AoA), and path gains are then estimated in a second processing stage by solving independent compressed sensing (CS) problems (one for each MS). In contrast to the classical approach in the literature, in which the DL and UL channel estimation problems are usually considered as two separate problems, our idea is to jointly estimate both the DL and UL channels as a single problem by concentrating most of the processing burden for channel estimation at the BS side. Simulation results demonstrate that the proposed receivers achieve a performance close to the classical approach that is applied on DL and UL communication links separately, with the advantage of avoiding complex computations for channel estimation at the MS side as well as dedicated feedback channels for each MS, which are attractive features for massive MIMO systems.

Applications of direction of arrival (DoA) techniques have dramatically increased in various areas ranging from the traditional wireless communication systems and rescue operations to GNSS systems and drone tracking. Particularly, police forces and security companies have drawn their attention to drone tracking devices, in order to provide the safeness of citizens and of clients, respectively. In this paper, we propose a low cost antenna array based drone tracking device for outdoor environments. The proposed solution is divided into hardware and software parts. The hardware part of the proposed device is based on off-the-shelf components such as an omnidirectional antenna array, a 4-channel software defined radio (SDR) platform with carrier frequency ranging from 70 MHz to 6 GHz, a FPGA motherboard, and a laptop. The software part includes algorithms for calibration, model order selection (MOS), and DoA estimation, including specific preprocessing steps and a tensor-based estimator to increase the DoA accuracy. We evaluate the performance of our proposed low cost solution in outdoor scenarios. According to our measurement campaigns, we show that when the array is in the front fire position, i.e., with a DoA ranging from -60° to 60° , the maximum and the average DoA errors are 6° and 1,9°, respectively.

With the widespread use of electronic devices such as laptops, smartphones and tablets, distance learning (DL) courses are increasingly being adopted by educational institutions. In order to conceive and execute such courses, different teaching models can be chosen. In this work, we overview the theoretical foundations related to the techniques of development of courses mediated by technologies from the Analysis, Design, Development, Implementation and Evaluation (ADDIE) model. We propose the extended ADDIE (X-ADDIE) for the elaboration of DL courses. In the proposed X-ADDIE, besides the execution step, a new step is added, defined as pilot experiment. All the X-ADDIE steps are tracked by a project management tool. We evaluate the proposed X-ADDIE model by means of a qualitative approach.

In this paper, the joint effect of multiple jamming signals and noise at the eavesdropper in wiretap channels is studied in terms of the secrecy outage probability, where all nodes are deployed with multiple antennas. Specifically, the transmitter employs a transmit antenna selection (TAS) technique; the legitimate receiver uses either a selection combining (SC) or a maximal-ratio combining (MRC) scheme; and the eavesdropper adopts the MRC technique. Exact closed-form expressions for the secrecy outage probability are derived for both SC and MRC schemes at the legitimate receiver, which are applicable for arbitrary number of power distributed jamming signals. An asymptotic secrecy outage analysis is conducted, which shows that the diversity gains for all combining schemes equal to the product of the number of antennas at the transmitter and the legitimate receiver. Representative numerical examples are presented to illustrate the effects of the key system parameters on the secrecy outage performance. Finally, the proposed analysis is validated through Monte Carlo simulation results.

Non‐orthogonal multiple access (NOMA) in amplify‐and‐forward relay systems with partial relay selection (PRS) is investigated. More specifically, new exact closed‐form expressions for the outage probabilities at two users are derived, based on which an asymptotic analysis at high signal‐to‐noise ratio (SNR) is carried out. Additionally, to investigate the performance gap between the NOMA and orthogonal multiple access (OMA) schemes, a closed‐form approximate expression at high SNR for the sum rate is derived. Furthermore, relying on the results, the impact of the PRS on the sum rate and outage probability of the proposed NOMA scheme is examined. In particular, the derived asymptotic expressions show that the proposed scheme can improve over the traditional OMA not only the sum rate but also the user fairness. Finally, simulation results are presented to corroborate the analytical results.

Amongst the several problems with the Brazilian e-government services (e-Gov), many of them are related to the fact that Brazil does not have a National Strategy for Digital Identity Management (IdM). To design a national strategy, it is crucial to analyze the solutions adopted in other countries and take into account Brazilian characteristics, such as its area, population, digital inclusion index, socio-political and economic profile, the current national identity registration system and the fraud rate. In this context, this paper aims to propose a national strategy for IdM for the Brazilian e-Gov program. This strategy was built upon by a literature review of the experiences of European countries and, mainly, the peculiarities inherent to Brazil.

Technological advances in information and communications systems are enabling access to education for millions of students, regardless of their geographical location. In this sense, distance learning schools take advantage of technological tools, such as course management systems (CMS) and their plugins. This work proposes a plugin for the e-learning platform Moodle that automatically sends e-mail messages to the students in order to motivate them to take part in the distance learning courses and, consequently, to reduce the dropout rates. This plugin has been tested in courses promoted by the General Coordination of Distance Education (CGEAD) of the Brazilian National School of Public Administration (ENAP). Based on data from its operation within the real ENAP distance learning environment, the proposed plugin is particularly successful to tackle the enrollment locking cases that compose the student dropout rate and also to motivate course enrolled students to access the virtual learning system more frequently.

In this paper, considering a mixed Rician and Rayleigh fading environment, we investigate the impact of line-of-sight (LOS) path component on the outage performance of dual-hop energy harvesting (EH) amplify-and-forward (AF) relay systems. To this end, a tight asymptotic outage lower bound expression is derived, which reveals that the system diversity order is limited to one and the outage behavior decays as log(SNR)/SNR, with SNR denoting the transmit signal-to-noise ratio (SNR) at the source. In contrast, when the LOS path component in the first-hop link becomes very strong (i.e., a large Rician K factor), our analytical results show that the easing-off factor log(SNR) can be effectively eliminated, becoming the decaying rate of outage curves steeper in this case and, consequently, improving the end-to-end transmission robustness since the outage curves scale as 1/SNR at high SNR regions.

In this paper, we investigate the effects of line-of-sight (LoS) path and opportunistic scheduling (OS) on dualhop energy harvesting (EH) relay systems, where the LoS path exists between source and EH relay. We first study a single-destination case and show that the system diversity order is one and the outage behavior decays as log(SNR)/SNR, with SNR denoting the transmit signal-to-noise ratio (SNR) at the source. In the presence of a strong LoS path, asymptotic results reveal that the easing-off factor log(SNR) can be eliminated and a faster decay rate, 1/SNR, is achieved. Afterward, the OS of the second hops (N destinations) is considered, and its separate as well as joint effects on the system outage behavior are examined. In addition, the choice of power-splitting factor is discussed and the impacts of EH constraint on the performance limit compared with conventional relay system are investigated, showing that under the same transmit-power consumption from the fixed energy supply, the EH relay system could achieve the same performance with that of conventional relay system if the wireless power transfer in the first hop is lossless and the energy conversion efficiency at EH relay tends to be unity.

In order to prevent the attack and discovery of sensitive data, a model is presented to protect the confidentiality of data even when the attackers have access to all data from the server: using systems that perform queries and inserts on encrypted data without the decryption key. The model set is applied in a real scenario: a distributed system hosted by Google (Encrypted BigQuery), which includes the use of encryption in data stored in non-relational databases for massive data processing. As a result, it is shown that this system supports a variety of applications with low overhead.

Wireless sensor networks (WSNs) are currently employed in a vast number of different applications ranging from home automation and health care to military systems. Although their application may vary greatly, WSNs share a common set of characteristics such as a limited energy supply and simple hardware. A common issue related with the application of WSNs is sensor localization, for some types of applications it is important that the sensors know the relative or absolute position of other sensors in the network, such as surveillance of monitoring networks. If sensors are randomly placed they may resort a wide range of methods such as Global Navigation Satellite Systems (GNSS) or received signal strength indicators (RSSI). In this work we present an alternative to relative sensor localization by employed a crossed dipole antenna in the reception and a known polarization in the transmission. The accuracy of the proposed methods is measured trough numerical simulations and results are presented.

Wireless sensor networks are evolving to hybridnetworks in which static and mobile sensor nodes cooperate in order to address challenging requirements imposed by new emerging applications. However, due to the ad hoc nature of the network and especially to resources constraints of the sensor nodes, this cooperation is not trivial, requiring a number of retransmissions thus wasting precious resources. In this paper the use of cooperative multiple input multiple output (MIMO) techniques is proposed to overcome transmission problems, ensuring a reliable and more efficient communication link with less retransmissions. Extensive simulation experiments support the proposal, and the results highlight the benefits in using MIMO to deliver messages from static to mobile nodes in wireless sensor networks.

To provide better life quality for mankind through the use of electrical and electronic devices that surround the world, controlling and communication between these equipments must occur and it must happen in an automatic and transparent way. This paper proposes a model based on usual technology to enable the event reading and the controlling of electrical devices, already existent and later developed, through a RESTful API or UPnP technology in order to allow the easy development of applications that can effectively improve life by transparent interaction with regular people. Furthermore, it will be proposed two direct applications of these APIs and it will be shown how to extend it.

Computing systems trust and reputation representation have been widely discussed and applied in a lot of information technology scenarios, becoming subject of scientific researches both from a theoretical and practical point of view. This paper proposes the development of a trust model to ensure a reliable files exchange among users in a private cloud, as well as the calculation process of trust among these users, according to the metrics previously established. The simulation result shows that this trust model is robust because it successfully access node reliable under Private Cloud Computing environment.

Frequently, R-dimensional subspace-based methods are used to estimate the parameters in multidimensional harmonic retrieval problems in a variety of signal processing applications. Since the measured data is multi-dimensional, traditional approaches require stacking the dimensions into one highly structured matrix. Recently, we have proposed eigenvalue based multi-dimensional model order selection schemes, which exploit the multidimensional structure of the data in order to achieve a higher probability of correct detection. However, our proposed multi-dimensional schemes are restricted to white noise scenarios. In this paper, we show how the Higher Order Singular Value Decomposition (HOSVD) of the measurement tensor enables us to improve the model order estimation of the estimation error (ESTER) scheme in colored noise scenarios. By using the matrix based ESTER, only one set of eigenvectors is taken into account, while by using the R-dimensional ESTER we obtain R sets of n-mode eigenvectors of the measurement tensor that are jointly used to improve significantly the accuracy of the estimated model order.

In This work we propose a new distributed intrusion detection system (IDS) designed for mobile ad hoc network (MANET) environments. The complete distribution of the intrusion detection process is the salient feature of our proposition: distribution is not restricted to data collection but also applied to execution of the detection algorithm and alert correlation. Each node in the MANET runs a local IDS (LIDS) that cooperates with others LIDS. A mobile agent framework is used to preserve the autonomy of each LIDS while providing a flexible technique for exploring the natural redundancies in MANET to compensate for the dynamic state of wireless links between high mobility nodes. The proposed solution has been validated by actual implementation, which is described in the paper. Three attacks are presented as illustrative examples of the IDS mechanisms. Attack detection is formally described by specification of data collection, attack signatures associated with such data and alerts generation and correlation. Experiments exhibit fairly good results, the attacks being collaboratively detected in real-time.