Manuel Rosa-Zurera

Detecting violent content in audio recordings is crucial for public safety, autonomous surveillance, and content moderation, particularly when visual cues are unreliable or unavailable. A resource-aware two-stage cascade system is proposed for acoustic violence detection that combines a lightweight Least Squares Linear Detector (LSLD) as a first-stage screener with a trimmed version of YAMNet as a second-stage classifier. A percentile-based forwarding rule controls the fraction of segments routed to the deep stage, turning the accuracy–cost trade-off into an explicit operating parameter for always-on deployment. The approach is evaluated on a publicly released dataset of real-world violent audio augmented with background noise and artificial reverberation. The results in the low-false-alarm regime show that the proposed cascade preserves performance close to a Stage 2-only baseline while substantially reducing average deep-inference workload. An ablation study validates the role of the LSLD as an inexpensive pre-filter, and robustness is assessed under clean, reverberant, and 12 dB noise conditions. Finally, an analytic energy consumption model is provided, which links computational workload to daily energy demand and photovoltaic sizing on ultra-low-power hardware, supporting sustainable off-grid deployment.

Background/Objectives: The implementation of artificial intelligence-based systems for disease detection using biomedical signals is challenging due to the limited availability of training data. This paper deals with the generation of synthetic EEG signals using deep learning-based models, to be used in future research for training Parkinson’s disease detection systems. Methods: Linear models, such as AR, MA, and ARMA, are often inadequate due to the inherent non-linearity of time series. To overcome this drawback, long short-term memory (LSTM) networks are proposed to learn long-term dependencies in non-linear EEG time series and subsequently generate synthetic signals to enhance the training of detection systems. To learn the forward and backward time dependencies in the EEG signals, a Bidirectional LSTM model has been implemented. The LSTM model was trained on the UC San Diego Resting State EEG Dataset, which includes samples from two groups: individuals with Parkinson’s disease and a healthy control group. Results: To determine the optimal number of cells in the model, we evaluated the mean squared error (MSE) and cross-correlation between the original and synthetic signals. This method was also applied to select the length of the hidden state vector. The number of hidden cells was set to 14, and the length of the hidden state vector for each cell was fixed at 4. Increasing these values did not improve MSE or cross-correlation and unnecessarily increased computational complexity. The proposed model’s performance was evaluated using the mean-squared error (MSE), Pearson’s correlation coefficient, and the power spectra of the synthetic and original signals, demonstrating the suitability of the proposed method for this application. Conclusions: The proposed model was compared to Autoregressive Moving Average (ARMA) models, demonstrating superior performance. This confirms that deep learning-based models, such as LSTM, are strong alternatives to statistical models like ARMA for handling non-linear, multifrequency, and non-stationary signals.

The implementation of artificial intelligence-based systems for disease detection using biomedical signals is challenging due to the limited availability of training data. The ability to synthetically augment training datasets is therefore crucial. This paper proposes using Long Short-Term Memory (LSTM) networks to learn long-term dependencies in non-linear time series, and subsequently employing the trained model to generate synthetic signals for improved training of detection systems. Linear models, such as AR, MA, or ARMA statistical models, are often inadequate due to the inherent non-linearity of the time series. The original data consist of Electroencephalogram (EEG) recordings.

Aircraft envelope expansion during the installation of new underwing stores presents significant challenges, particularly due to the aeroelastic flutter phenomenon. Accurate modeling of aeroelastic behavior often necessitates flight testing, which poses risks due to the potential catastrophic consequences of reaching the flutter point. Traditional methods, like frequency sweeps, are effective but require prolonged exposure to flutter conditions, making them less suitable for transonic flight validations. This paper introduces a robust deep learning approach to process sine dwell signals from aeroelastic flutter flight tests, characterized by short data lengths (less than 5 s) and low frequencies (less than 10 Hz). We explore the preliminary viability of different deep learning networks and compare their performances to existing methods such as the PRESTO algorithm and Laplace Wavelet Matching Pursuit estimation. Deep learning algorithms demonstrate substantial accuracy and robustness, providing reliable parameter identification for flutter analysis while significantly reducing the time spent near flutter conditions. Although the results with the networks trained show less accuracy than the PRESTO algorithm, they are more accurate than the Laplace Wavelet estimation, and the results are promising enough to justify extended investigation on this area. This approach is validated using both synthetic data and real F-18 flight test signals, which highlights its potential for real-time analysis and broader applicability in aeroelastic testing.

Aircraft envelope expansion during the installation of new underwing stores presents significant challenges, particularly due to the aeroelastic flutter phenomenon. Accurate modeling of aeroelastic behavior often necessitates flight testing, which poses risks due to the potential catastrophic consequences of reaching the flutter point. Traditional methods, like frequency sweeps, are effective but require prolonged exposure to flutter conditions, making them less suitable for transonic flight validations. This paper introduces a robust deep learning approach to process sine-dwell signals from aeroelastic Flutter Flight Tests, characterized by short data lengths (less than 5 seconds) and low frequencies (less than 10 Hz). We explore the preliminary viability of different deep learning networks and compare their performance to existing methods such as the PRESTO algorithm and Laplace Wavelet estimation. Deep learning algorithms demonstrate substantial accuracy and robustness, providing reliable parameter identification for flutter analysis while significantly reducing the time spent near flutter conditions. Although the results with the networks trained show less accuracy than the PRESTO algorithm but more than the Laplace Wavelet estimation, the results are promising enough to justify extended investigation on this area. This approach is validated using both synthetic data and real F-18 flight test signals, highlighting its potential for real-time analysis and broader applicability in aeroelastic testing.