Optical and transport properties of wurtzite GaN layers, grown by plasma-assisted molecular-beam epitaxy on Si(111) substrates, have been investigated. An emission at 3.455 eV, analyzed by continuous-wave and time-resolved luminescence in undoped and Si-doped GaN layers, is assigned to excitons bound to Si donors with an optical binding energy of 50 meV. A common origin of this peak, for undoped and Si-doped GaN, is backed by secondary-ion-mass spectroscopy that evidences a Si diffusion from the substrate into the GaN layer for growth temperatures above 660 °C. Simultaneously, Ga diffusion into the Si substrate generates a highly $p$-type conductive layer at the GaN/Si interface, leading to unreliable Hall data in undoped and lightly doped layers. Positron annihilation reveals a concomitant vacancy cluster generation at the GaN/Si interface in samples grown above 660 °C. No traces of the “yellow band” are detected either in undoped or in Si-doped samples.

Deep learning (DL) models are very useful for human activity recognition (HAR); these methods present better accuracy for HAR when compared to traditional, among other advantages. DL learns from unlabeled data and extracts features from raw data, as for the case of time-series acceleration. Sliding windows is a feature extraction technique. When used for preprocessing time-series data, it provides an improvement in accuracy, latency, and cost of processing. The time and cost of preprocessing can be beneficial especially if the window size is small, but how small can this window be to keep good accuracy? The objective of this research was to analyze the performance of four DL models: a simple deep neural network (DNN); a convolutional neural network (CNN); a long short-term memory network (LSTM); and a hybrid model (CNN-LSTM), when variating the sliding window size using fixed overlapped windows to identify an optimal window size for HAR. We compare the effects in two acceleration sources’: wearable inertial measurement unit sensors (IMU) and motion caption systems (MOCAP). Moreover, short sliding windows of sizes 5, 10, 15, 20, and 25 frames to long ones of sizes 50, 75, 100, and 200 frames were compared. The models were fed using raw acceleration data acquired in experimental conditions for three activities: walking, sit-to-stand, and squatting. Results show that the most optimal window is from 20–25 frames (0.20–0.25s) for both sources, providing an accuracy of 99,07% and F1-score of 87,08% in the (CNN-LSTM) using the wearable sensors data, and accuracy of 98,8% and F1-score of 82,80% using MOCAP data; similar accurate results were obtained with the LSTM model. There is almost no difference in accuracy in larger frames (100, 200). However, smaller windows present a decrease in the F1-score. In regard to inference time, data with a sliding window of 20 frames can be preprocessed around 4x (LSTM) and 2x (CNN-LSTM) times faster than data using 100 frames.

Peer reviewed article authored by (Fernandez Cabada T, Sanchez Lopez de Pablo C, Martinez Serrano A, del Pozo Guerrero F, Serrano Olmedo JJ, Ramos Gomez M). Read article or submit your manuscript for publishing.

Human activity recognition (HAR) has applications ranging from security to healthcare. Typically these systems are composed of data acquisition and activity recognition models. In this work, we compared the accuracy of two acquisition systems: Inertial Measurement Units (IMUs) vs Movement Analysis Systems (MAS). We trained models to recognize arm exercises using state-of-the-art deep learning architectures and compared their accuracy. MAS uses a camera array and reflective markers. IMU uses accelerometers, gyroscopes, and magnetometers. Sensors of both systems were attached to different locations of the upper limb. We captured and annotated 3 datasets, each one using both systems simultaneously. For activity recognition, we trained 8 architectures, each one with different operations and layers configurations. The best architectures were a combination of CNN, LSTM, and Transformer achieving test accuracy from 89% to 99% on average. We evaluated how feature selection reduced the sensors required. We found IMU and MAS data were able to distinguish correctly the arm exercises. CNN layers at the beginning produced better accuracy on challenging datasets. IMU had advantages over other aquisition systems for activity recognition. We analyzed the relations between models accuracy, signal waveforms, signals correlation, sampling rate, exercise duration, and window size. Finally, we proposed the use of a single IMU located at the wrist and a variable-size window extraction.

For decades now, conventional sinusoidal signals have been exclusively used in magnetic hyperthermia as the only alternating magnetic field waveform to excite magnetic nanoparticles. However, there are no theoretical nor experimental reasons that prevent the use of different waveforms. The only justifiable motive behind using the sinusoidal signal is its availability and the facility to produce it. Following the development of a configurable alternating magnetic field generator, we aim to study the effect of various waveforms on the heat production effectiveness of magnetic nanoparticles, seeking to prove that signals with more significant slope values, such as the trapezoidal and almost-square signals, allow the nanoparticles to reach higher efficiency in heat generation. Furthermore, we seek to point out that the nanoparticle power dissipation is dependent on the waveform’s slope and not only the frequency, magnetic field intensity and the nanoparticle size. The experimental results showed a remarkably higher heat production performance of the nanoparticles when exposed to trapezoidal and almost-square signals than conventional sinusoidal signals. We conclude that the nanoparticles respond better to the trapezoidal and almost-square signals. On the other hand, the experimental results were used to calculate the normalized power dissipation value and prove its dependency on the slope. However, adjustments are necessary to the coil before proceeding with in vitro and in vivo studies to handle the magnetic fields required.