In this paper, a large-area flexible tactile sensor for multi-touch and force detection based on EIT technology was developed. A novel design of a sensor material made of a porous elastic polymer and ionic liquid was proposed. The proposed conductive flexible materials combining elastic porous structures and conductive liquids provide continuous, linear changes in impedance with respect to touch forces. A deep learning scheme PSPNet based on MobileNet was adopted to postprocess the originally reconstructed images to improve the performance of tactile perception. By using this data-driven method, we can improve the spatial resolution of the tactile sensor to achieve a single-point position detection error of 7.5±4.5 mm without using internal electrodes.

Efficient and realistic indoor radio propagation modelling tools are inextricably intertwined with the design and operation of next generation wireless networks. Machine learning (ML)-based radio propagation models can be trained with simulated or real-world data to provide accurate estimates of the wireless channel characteristics in a computationally efficient way. However, most of the existing research on ML-based propagation models focuses on outdoor propagation modelling, while indoor data-driven propagation models remain site-specific with limited scalability. In this paper we present an efficient and credible ML-based radio propagation modelling framework for indoor environments. Specifically, we demonstrate how a convolutional encoder-decoder can be trained to replicate the results of a ray-tracer, by encoding physics-based information of an indoor environment, such as the permittivity of the walls, and decode it as the path-loss (PL) heatmap for an environment of interest. Our model is trained over multiple indoor geometries and frequency bands, and it can eventually predict the PL for unknown indoor geometries and frequency bands within a few milliseconds. Additionally, we illustrate how the concept of transfer learning can be leveraged to calibrate our model by adjusting its pre-estimate weights, allowing it to make predictions that are consistent with measurement data.

Comparative study of the performance of different ITU model in predicting Malaysia tropical climate properties.

The preprint has been already submitted to an IEEE journal for peer review.

Design adaptive equalizer using Affine projection algorithm for MIMO SC-FDMA and compare it with MMSE . Both equalizers are used within overlap-save method. Also a turbo decoder is used in conjunction with overlap-save method to enhance the BER performances in 8x8 and 16x16 MIMO SC-FDMA.

Deep-learning models estimate values using backpropagation. The activation function within hidden layers is a critical component to minimizing loss in deep neural-networks. Rectified Linear (ReLU) has been the dominant activation function for the past decade. Swish and Mish are newer activation functions that have shown to yield better results than ReLU given specific circumstances. Phish is a novel non-monotonic activation function proposed here. It is a composite function defined as f(x) = xTanH(GELU(x)), where no discontinuities are apparent in the differentiated graph on the domain observed. Four generalized networks were constructed using different activation functions. SoftMax was the output function. Using images from MNIST and CIFAR-10 databanks, these networks were trained to minimize sparse categorical crossentropy. A large-scale cross-validation was simulated using stochastic Markov chains to account for the law of large numbers for the probability values. Statistical tests support the research hypothesis stating Phish could outperform other activation functions in classification. Future experiments would involve testing Phish in unsupervised learning algorithms and comparing it to more activation functions.

Deep learning (DL) models have significantly improved the performance of text classification and text regression tasks. However, DL models are often strikingly vulnerable to adversarial attacks. Many researchers have aimed to develop adversarial attacks against DL models in realistic black-box settings (i.e., assumes no model knowledge is accessible to attackers) that typically operate with a two-phase framework: (1) sensitivity estimation through gradient-based or deletion-based methods to evaluate the sensitivity of each token to the prediction of the target model, and (2) perturbation execution to craft adversarial examples based on the estimated token sensitivity. However, gradient-based and deletion-based methods used to estimate sensitivity often face issues of capturing the directionality of tokens and overlapping token sensitivities, respectively. In this study, we propose a novel eXplanation-based method for Adversarial Text Attacks (XATA) that leverages additive feature attribution explainable methods, namely LIME or SHAP, to measure the sensitivity of input tokens when crafting black-box adversarial attacks on DL models performing text classification or text regression. We evaluated XATA’s attack performance on DL models executing text classification on three datasets (IMDB Movie Review, Yelp Reviews-Polarity, and Amazon Reviews-Polarity) and DL models conducting text regression on three datasets (My Personality, Drug Review, and CommonLit Readability). The proposed XATA outperformed the existing gradient-based and deletion-based adversarial attack baselines in both tasks. These findings indicate that the ever-growing research focused on improving the explainability of DL models with additive feature attribution explainable methods can provide attackers with weapons to launch targeted adversarial attacks.

This paper addresses probabilistic shaping (PS) which has been a latest key technique to approach capacity of fiber-optic channels. We investigate the impact of PS on nonlinear interference (NLI), including self channel interference (SCI), cross channel interference (XCI), and multi channel interference (MCI) for a polarization multiplexed 16-ary quadrature amplitude modulation format in a wavelength division multiplexed (WDM) system. To this end, we consider performing PS in two scenarios: (i) Solely on the channel of interest and (ii) over all C-band WDM channels of a fiber-optic link by analyzing the effective signal to noise ratio and symbol error rate. It is demonstrated that using the enhanced Gaussian noise model with merely 10% overhead in the first scenario, the applied PS scheme increases the SCI and the total experienced NLI by about 19.23%, and 6.6%, respectively. Interestingly, despite enhancing the NLI in this scenario, the simulated PS technique leads to about 47.6% increase in the transmission reach. In the second scenario, the numerical results show increase of the SCI, XCI, and total NLI around 19.8%, 23.34%, and 20.2%, respectively, but resulting in an increase of 32.3% in the transmission reach.

Network and system security are incredibly critical issues now. Due to the rapid proliferation of malware, traditional analysis methods struggle with enormous samples. In this paper, we propose four easy-to-extract and small-scale features, including sizes and permissions of Windows PE sections, content complexity, and import libraries, to classify malware families, and use automatic machine learning to search for the best model and hyper-parameters for each feature and their combinations. Compared with detailed behavior-related features like API sequences, proposed features provide macroscopic information about malware. The analysis is based on static disassembly scripts and hexadecimal machine code. Unlike dynamic behavior analysis, static analysis is resource-efficient and offers complete code coverage, but is vulnerable to code obfuscation and encryption. The results demonstrate that features which work well in dynamic analysis are not necessarily effective when applied to static analysis. For instance, API 4-grams only achieve 57.96% accuracy and involve a relatively high dimensional feature set (5000 dimensions). In contrast, the novel proposed features together with a classical machine learning algorithm (Random Forest) presents very good accuracy at 99.40% and the feature vector is of much smaller dimension (40 dimensions). We demonstrate the effectiveness of this approach through integration in IDA Pro, which also facilitates the collection of new training samples and subsequent model retraining.

The authors have developed a prototype, which promises easy passage for trains in unknown environments such as zero visibility situations, fog, smog, and heavy rain. The prototype is based upon the integration of a camera, Radio Detection and Ranging (RADAR), and Infrared (IR) Light Amplification by Stimulated Emission of Radiation (LASER).The camera used in the prototype catches a long-range view of the track and presents it live on a mini screen fixed in the loco pilot cabin. A combined short, mid, and long-range radar sensor system is used to detect obstacles continuously in loco pilot blind spots on the track, particularly for collision avoidance assistance at high speed. The present work proposes a long-range laser IR illuminator with a wide range of color and mono cameras to aid clear and precise monitoring in zero visibility conditions, which is fixed on the front portion/engine of the locomotive/engine. The prototype experimental results for 2 m - 2 km distances have been performed on a live running train, which shows that the developed prototype tracks obstacles effectively during fog and smog conditions. The design concept, observation, prototype model, and other technical specifications have been presented, and satisfactory results were found.

A new circuit of the output stage of an operational amplifier implemented on GaAs n-channel field-effect transistors with a control p-n junction and GaAs bipolar p-n-p transistors is investigated. Its peculiarity consists in the presence of a nonlinear negative feedback that stabilizes the drain current of the output transistor with an n-channel at a negative input voltage. The basic equations for the static mode of the output stage are given. The results of modeling in the LTspice simulation software of 3 modifications of the proposed circuit solutions are discussed.

Preprint of our study “Robust Multisensor Prediction of Drought-Induced Soybean Yield Anomalies in Argentina”

Muzzle voltage is an essential diagnostic tool used in both contact resistance modeling and transition determination. However, it is challenging to stem the necessary meanings from the collected measurements. In this study, EMFY-3 launch experiments are used to model muzzle voltage characteristics to understand the transition mechanism better. These experiments have muzzle energies in the range between 1.69-2.85 MJ in ASELSAN Electromagnetic Launcher Laboratory. Six different launch tests with various rail current waveforms that ranged between 1.5-2.1 MA are used to investigate different scenarios. Some parameters which affect muzzle voltage are calculated with the 3-D Finite Element Method (FEM), i.e., rail mutual inductance $\mathrm{L_m}$. Muzzle voltages are decomposed into subsections; each subsection is calculated with proper models. Simulation results are coherent with experimental measurements. Findings are compared with previous studies, and differences are explained with possible reasons. Even though we could not conclusively resolve which physical quantity starts to transition, the study showed that transition does not form a specific muzzle velocity, armature action integral, or down-slope rail current ratio.

Particle swarm optimization system based on the distributed architecture has shown its efficiency for static optimization and has not been studied to solve dynamic multiobjective problems (DMOPs). When solving DMOP, tracking the best solutions over time and ensuring good exploitation and exploration are the main challenging tasks. This study proposes a novel Dynamic Pareto bi-level Multi-Objective Particle Swarm Optimization (DPb-MOPSO) algorithm including two parallel optimization levels. At the first level, all solutions are managed in a single search space. When a dynamic change is successfully detected, the Pareto ranking operator is used to enable a multiswarm subdivisions and processing which drives the second level of enhanced exploitation. A dynamic handling strategy based on random detectors is used to track the changes of the objective function due to time-varying parameters. A response strategy consisting in re-evaluate all unimproved solutions and replacing them with newly generated ones is also implemented. Inverted generational distance, mean inverted generational distance, and hypervolume difference metrics are used to assess the DPb-MOPSO performances. All quantitative results are analyzed using Friedman’s analysis while the Lyapunov theorem is used for stability analysis. Compared with several multi-objective evolutionary algorithms, the DPb-MOPSO is robust in solving 21 complex problems over a range of changes in both the Pareto optimal set and Pareto optimal front. For 13 UDF and ZJZ functions, DPb-MOPSO can solve 8/13 and 7/13 on IGD and HVD with moderate changes. For the 8 FDA and dMOP benchmarks, DPb-MOPSO was able to resolve 4/8 with severe change on MIGD, and 5/8 for moderate and slight changes. However, for the 3 kind of environmental changes, DPb-MOPSO assumes 4/8 of the solving function on IGD and HVD.

KITTI dataset is collected from three types of environments, i.e., country, urban and highway The types of feature point cover a variety of scenes. The KITTI dataset provides 22 sequences of LiDAR data. 11 sequences of them from sequence 00 to sequence 10 are “training” data. The training data are provided with ground truth translation and rotation. In addition, field experiment data is collected by low-resolution LiDAR, VLP-16 in Wuhan Research and Innovation Center.

In this paper was presented Itamaracá, a novel simple way to generate pseudo random numbers. In general vision we can say that Itamaracá tends to pass in some statistical tests like frequency, chi square, autocorrelation, run sequence and run test. As an effect to comparison also was taking into account the results of the function R and Between by Microsoft Excel and true random numbers by Random Org analyzed its distinctive characteristics as well as with the proposal model. In this sense, the goal of this study is contributing to growing the existing Pseudo Random Number Generators (PRNGs) portfolio.

Observing decline in machine performance using a Linear Regression model

We propose a novel hybrid framework for registering retinal images in the presence of extreme geometric distortions that are commonly encountered in ultra-widefield (UWF) fluorescein angiography. Our approach consists of two stages: a feature-based global registration and a vessel-based local refinement. For the global registration, we introduce a modified RANSAC algorithm that jointly identifies robust matches between feature keypoints in reference and target images and estimates a polynomial geometric transformation consistent with the identified correspondences. Our RANSAC modification particularly improves feature point matching and the registration in peripheral regions that are most severely impacted by the geometric distortions. The second local refinement stage is formulated in our framework as a parametric chamfer alignment for vessel maps obtained using a deep neural network. Because the complete vessel maps contribute to the chamfer alignment, this approach not only improves registration accuracy but also aligns with clinical practice, where vessels are typically a key focus of examinations. We validate the effectiveness of the proposed framework on a new UWF fluorescein angiography (FA) dataset and on the existing narrow-field FIRE (fundus image registration) dataset and demonstrate that it significantly outperforms prior retinal image registration methods. The proposed approach enhances the utility of large sets of longitudinal UWF images by enabling: (a) automatic computation of vessel change metrics and (b) standardized and co-registered examination that can better highlight changes of clinical interest to physicians.