Search Results (6)

In recent years, the research on human activity recognition based on channel state information (CSI) of Wi-Fi has gradually attracted much attention in order to avoid the deployment of additional devices and reduce the risk of personal privacy leakage. In this paper, we propose a hybrid network architecture, named VBCNet, that can effectively identify human activity postures. Firstly, we extract CSI sequences from each antenna of Wi-Fi signals, and the data are preprocessed and tokenised. Then, in the encoder part of the model, we introduce a layer of long short-term memory network to further extract the temporal features in the sequences and enhance the ability of the model to capture the temporal information. Meanwhile, VBCNet employs a convolutional feed-forward network instead of the traditional feed-forward network to enhance the model’s ability to process local and multi-scale features. Finally, the model classifies the extracted features into human behaviours through a classification layer. To validate the effectiveness of VBCNet, we conducted experimental evaluations on the classical human activity recognition datasets UT-HAR and Widar3.0 and achieved an accuracy of 98.65% and 77.92%. These results show that VBCNet exhibits extremely high effectiveness and robustness in human activity recognition tasks in complex scenarios. Full article

An increased jumping performance is key for gymnastics competition routines. Rhythmic gymnasts (RGs) use the jump as one of the main body elements. In Acrobatic Gymnastics (ACRO), top gymnasts must coordinate their jumps with the impulse provided by base gymnast(s). It is expected that the gymnasts’ discipline and role played impact their jumping skill. This work aims to investigate how the jumping performance differs between ACRO gymnasts and RGs, focusing on the Force–Velocity (F-V) profile mechanical variables. Gymnasts were divided in three groups: ACRO tops (n = 10, 13.89 (3.62) median (interquartile interval) years old), ACRO bases (n = 18, 18.24 (4.41) years old) and RGs (n = 15, 12.00 (3.00) years old). The F-V profile during countermovement jump and its mechanical variables were evaluated using MyJump2. A training background survey and anthropometric assessments were conducted. The significance level was set at p ≤ 0.05. Group comparisons showed that ACRO bases jump higher than ACRO tops and RGs, present a higher maximal force than RGs and a more balanced F-V profile, while RGs present high force deficits. Coaches can use this data to develop interventions that optimize the training stimulus to different gymnastics disciplines considering the individual characteristics and adaptability of each gymnast. Full article

The past decade has demonstrated the potential of human activity recognition (HAR) with WiFi signals owing to non-invasiveness and ubiquity. Previous research has largely concentrated on enhancing precision through sophisticated models. However, the complexity of recognition tasks has been largely neglected. Thus, the performance of the HAR system is markedly diminished when tasked with increasing complexities, such as a larger classification number, the confusion of similar actions, and signal distortion To address this issue, we eliminated conventional convolutional and recurrent backbones and proposed WiTransformer, a novel tactic based on pure Transformers. Nevertheless, Transformer-like models are typically suited to large-scale datasets as pretraining models, according to the experience of the Vision Transformer. Therefore, we adopted the Body-coordinate Velocity Profile, a cross-domain WiFi signal feature derived from the channel state information, to reduce the threshold of the Transformers. Based on this, we propose two modified transformer architectures, united spatiotemporal Transformer (UST) and separated spatiotemporal Transformer (SST) to realize WiFi-based human gesture recognition models with task robustness. SST intuitively extracts spatial and temporal data features using two encoders, respectively. By contrast, UST can extract the same three-dimensional features with only a one-dimensional encoder, owing to its well-designed structure. We evaluated SST and UST on four designed task datasets (TDSs) with varying task complexities. The experimental results demonstrate that UST has achieved recognition accuracy of 86.16% on the most complex task dataset TDSs-22, outperforming the other popular backbones. Simultaneously, the accuracy decreases by at most 3.18% when the task complexity increases from TDSs-6 to TDSs-22, which is 0.14–0.2 times that of others. However, as predicted and analyzed, SST fails because of excessive lack of inductive bias and the limited scale of the training data. Full article

The unsteady flow of biological liquid through non-uniform pumps under porosity impacts is considered. The Jeffrey fluid is used as blood in the current study, which is also characterized as viscoelastic fluid because of its dual characteristics: on the one hand, its viscosity in nature; on the other hand, its elastic effect. Rheological equations are framed in a curvilinear coordinates system, and porosity influences are simulated with the body force term in momentum equations. The flow system has been transformed from fixed to wave frame using a linear–mathematical transformation between these two frames. In the next mathematical steps, these transformed equations are given in non-dimensional form using physical variables. The system of PDE is reduced to an ODE under lubrication theory and long wavelength approximation. Solutions to reduced ordinary differential equations are obtained numerically in MATLAB software via a BVP4C scheme. The physical impacts of the involved parameters on flow features, such as curvature, porosity (Darcy’s number), non-uniformity, and viscoelastic parameters, have been visualized graphically. Multi-sinusoidal waves are used in the boundary wall of the curved pump for peristaltic pumping. The magnitude of velocity profile for a saw-tooth wave (trapezoidal wave) is larger (smaller) than all other natures of peristaltic waves. The larger intensity of Darcy’s number has a dynamic role in the reduction of peristaltic pumping, whereas the opposite behavior is noticed when increasing the non-uniform nature of a channel. A comparison between all multi-sinusoidal waves is also addressed. The results of the present research shall be very productive for the manufacture of peristaltic pumps for drug delivery and bio-medical systems. Full article

Airfoil selection procedure, wind tunnel testing and an implementation of 6-DOF model on flying wing micro aerial vehicle (FWMAV) has been proposed in this research. The selection procedure of airfoil has been developed by considering parameters related to aerodynamic efficiency and flight stability. Airfoil aerodynamic parameters have been calculated using a potential flow solver for ten candidate airfoils. Eppler-387 proved to be the most efficient reflexed airfoil and therefore was selected for fabrication and further flight testing of vehicle. Elevon control surfaces have been designed and evaluated for longitudinal and lateral control. The vehicle was fabricated using hot wire machine with EPP styrofoam of density 50 Kg/ ${\mathrm{m}}^3$ . Static aerodynamic coefficients were evaluated using wind tunnel tests conducted at cruise velocity of 20 m/s for varying angles of attack. Rate derivatives and elevon control derivatives have also been calculated. Equations of motion for FWMAV have been written in a body axis system yielding a 6-DOF model. It was found during flight tests that vehicle conducted coordinated turns with no appreciable adverse yaw. Since FWMAV was not designed with a vertical stabilizer and rudder control surface, directional stability was therefore augmented through winglets and high wing leading edge sweep. Major problems encountered during flight tests were related to left rolling tendency. The left roll tendency was found inherent to clockwise rotating propeller as ‘P’ factor, gyroscopic precession, torque effect and spiraling slipstream. To achieve successful flights, many actions were required including removal of excessive play from elevon control rods, active actuation of control surfaces, enhanced launch speed during take off, and increased throttle control during initial phase of flight. FWMAV flew many successful stable flights in which intended mission profile was accomplished, thereby validating the proposed airfoil selection procedure, modeling technique and proposed design. Full article

A theoretical and a mathematical model is presented to determine the entropy generation on electro-kinetically modulated peristaltic propulsion on the magnetized nanofluid flow through a microchannel with joule heating. The mathematical modeling is based on the energy, momentum, continuity, and entropy equation in the Cartesian coordinate system. The effects of viscous dissipation, heat absorption, magnetic field, and electrokinetic body force are also taken into account. The electric field terms are helpful to model the electrical potential terms by means of Poisson–Boltzmann equations, ionic Nernst–Planck equation, and Debye length approximation. A perturbation method has been applied to solve the coupled nonlinear partial differential equations and a series solution is obtained up to second order. The physical behavior of all the governing parameters is discussed for pressure rise, velocity profile, entropy profile, and temperature profile. Full article