Eduardo Parente Ribeiro

Forensic sketching translates a witness description into a visual representation of a suspect, but sketches often lack visual details, which can limit their use in face recognition systems. This paper presents a sketch-guided face generation pipeline based on a Conditional Variational Autoencoder (CVAE) designed to generate facial reconstructions from sketches and conditional attributes. The method uses a stochastic preprocessing pipeline to extract edge maps from facial photographs, reducing dependence on manually paired sketch-photo datasets. Conditional inputs are incorporated to control ambiguous attributes that may not be fully specified by the sketch. The proposed approach is evaluated using synthetic edge maps, hand-drawn sketches, and digitally drawn sketches from different sources, considering both reconstruction quality and identity-oriented similarity. Compared with a vanilla autoencoder, the proposed CVAE reduced the FaceNet distance by 56.8% in the CUHK reconstruction evaluation. The experimental results also suggest that considering both reconstruction quality and identity preservation may provide a more complete evaluation of sketch-guided face generation in forensic scenarios.

Electroencephalography (EEG) provides valuable information about a subject’s mental state; however, developing reliable classification models remains challenging. One major difficulty lies in defining an effective feature representation, as the wide range of features proposed in the literature often leads to high-dimensional inputs, increasing the risk of overfitting, reducing generalization, and raising computational cost. A further critical challenge is the strong inter-subject variability inherent to EEG data, where distributional shifts frequently cause models trained on one individual to perform poorly on unseen subjects. This work proposes a novel family of correlation-based feature selection methods that explicitly models inter-feature relationships through correlation structures. The objective is to identify features that are simultaneously discriminative across mental states (relaxed and concentrated) and invariant across subjects, thereby improving cross-subject generalization. The proposed methods are evaluated against established feature selection and dimensionality reduction techniques using a leave-one-subject-out experimental protocol, in which models are trained on multiple participants and tested on unseen individuals. Experimental results demonstrate that the proposed approach consistently achieves superior or competitive performance compared to existing methods, particularly under strong inter-subject distribution shifts. In addition, the analysis reveals how preprocessing parameters—such as window length, overlap, and frequency band decomposition—affect classification performance and generalization. Unlike previous EEG feature selection approaches that primarily focus on feature relevance or redundancy, the proposed framework explicitly promotes domain invariance while preserving feature interpretability, without relying on subject-specific calibration.

Airspace control plays an important role in the safety and fluidity of air traffic. A fundamental service for this purpose is audio communication through frequencies in the VHF bands. This paper describes the evaluation of the audio degradation of voice transmissions from control centre to the aircraft. The effects of more than one station broadcasting on the same frequency with carrier offset (climax mode) are analysed using perceptual evaluation of speech quality (PESQ) perceptual model. Comparative studies are performed to verify the degree of degradation of different audio transport systems and climax situation.

The measurement of some physical properties requires detection of both amplitude and phase of an applied test signal. In portable sensors, this needs to be performed locally with integrated analog circuits or digital processing. Wireless sensor network allows the measurements over wide areas but raises a challenge of using efficient algorithm for phase sensitive detection (PSD) in nodes with constrained processing capacity. It is not straightforward to compare different PSD methods because their performance is closely related to the fine tuning of their parameters such as filter order and cut-off frequency. We propose a methodology to compare several PSD methods subjected to the same response time to assess its ability to provide accurate estimation in the presence of noise. We also determine the computational complexity of the investigated methods in terms of the number of operations required. We investigate both continuous and windowed operation. The methods were tested through simulations and verified using a dedicated embedded system. Our findings demonstrate that Goertzel algorithm rendered the best results, with smaller estimation error and less computation complexity.

The ability to detect elephant flows in the forwarding device itself, i.e., a switch, facilitates the deployment of new advanced applications such as load-balancing, per-flow QoS management, etc. Sketches and Space Saving summarization techniques are used for elephant flow detection. However, their memory and computing requirements force the cooperation of an external controller device, due to the scarce resources of current programmable switches. To overcome this limitation, we adapt Sketch and Space Saving elephant flow detection techniques to operate with instant notification and sampled traffic. We evaluate the performance of the resulting techniques with three real traffic traces. The use of sampling allows the identification of a large share of the total traffic corresponding to the elephant flows with a low memory footprint and a reduction of the computing requirements in two orders of magnitude compared to unsampled versions. In turn, we observe a slight increase in the number of false positives and the number of flow notifications.

The study investigates the self‐equalization effect and its impact on the spectral efficiency (SE) analysis in a massive multiple‐input multiple‐output (MIMO) scenario for 5G wireless networks for filter bank multicarrier (FBMC) and orthogonal frequency division multiplexing (OFDM) modulations. For that the SNR adopted in previous studies is compared to the inverse of the mean squared error (MSE) for multiple users in a single cell environment. In order to incorporate other distortions not correlated to the signal, this work uses the inverse of the MSE as the signal‐to‐noise‐and‐distortion ratio (SNDR) and, that metric is adopted for SE calculation in massive MIMO (MM) uplink setup. Throughout the article, the importance of considering this metric for the SE analysis of the MM system will be noticeable. Previous works dealing with MM focus on the case when the number of antennas is so much higher than the number of users that specific conditions could be applied, however, it was shown throughout this study that some of those assumptions could represent an imprecision, especially for the case when the ratio between the number of antennas and the number of users is less expressive.

In this work, we show the success of unsupervised transfer learning between Electroencephalographic (brainwave) classification and Electromyographic (muscular wave) domains with both MLP and CNN methods. To achieve this, signals are measured from both the brain and forearm muscles and EMG data is gathered from a 4-class gesture classification experiment via the Myo Armband, and a 3-class mental state EEG dataset is acquired via the Muse EEG Headband. A hyperheuristic multi-objective evolutionary search method is used to find the best network hyperparameters. We then use this optimised topology of deep neural network to classify both EMG and EEG signals, attaining results of 84.76% and 62.37% accuracy, respectively. Next, when pre-trained weights from the EMG classification model are used for initial distribution rather than random weight initialisation for EEG classification, 93.82%(+29.95) accuracy is reached. When EEG pre-trained weights are used for initial weight distribution for EMG, 85.12% (+0.36) accuracy is achieved. When the EMG network attempts to classify EEG, it outperforms the EEG network even without any training (+30.25% to 82.39% at epoch 0), and similarly the EEG network attempting to classify EMG data outperforms the EMG network (+2.38% at epoch 0). All transfer networks achieve higher pre-training abilities, curves, and asymptotes, indicating that knowledge transfer is possible between the two signal domains. In a second experiment with CNN transfer learning, the same datasets are projected as 2D images and the same learning process is carried out. In the CNN experiment, EMG to EEG transfer learning is found to be successful but not vice-versa, although EEG to EMG transfer learning did exhibit a higher starting classification accuracy. The significance of this work is due to the successful transfer of ability between models trained on two different biological signal domains, reducing the need for building more computationally complex models in future research.

In this study, we present a transfer learning method for gesture classification via an inductive and supervised transductive approach with an electromyographic dataset gathered via the Myo armband. A ternary gesture classification problem is presented by states of ’thumbs up’, ’thumbs down’, and ’relax’ in order to communicate in the affirmative or negative in a non-verbal fashion to a machine. Of the nine statistical learning paradigms benchmarked over 10-fold cross validation (with three methods of feature selection), an ensemble of Random Forest and Support Vector Machine through voting achieves the best score of 91.74% with a rule-based feature selection method. When new subjects are considered, this machine learning approach fails to generalise new data, and thus the processes of Inductive and Supervised Transductive Transfer Learning are introduced with a short calibration exercise (15 s). Failure of generalisation shows that 5 s of data per-class is the strongest for classification (versus one through seven seconds) with only an accuracy of 55%, but when a short 5 s per class calibration task is introduced via the suggested transfer method, a Random Forest can then classify unseen data from the calibrated subject at an accuracy of around 97%, outperforming the 83% accuracy boasted by the proprietary Myo system. Finally, a preliminary application is presented through social interaction with a humanoid Pepper robot, where the use of our approach and a most-common-class metaclassifier achieves 100% accuracy for all trials of a ‘20 Questions’ game.