This article proposes a wavelet-aided orthogonal time-frequency space (W-OTFS) modulation as a novel approach for high-mobility vehicular communication. Leveraging the advantages of discrete wavelet transforms, W-OTFS provides improved performance over existing orthogonal time-frequency space (OTFS) modulation in doubly selective vehicular channels. The system exhibits a direct correspondence between input and output through matrix multiplication and primarily employs a minimum mean square error (MMSE) based linear detector for signal detection. This study investigates Haar and Daubechies wavelets for the transformation and we observe that wavelets of lower-order enhance channel sparsity and reduce computational complexity. A soft-decision-based iterative detector is proposed to improve the system’s performance further. The results of our simulations demonstrate that W-OTFS utilizing the soft-decision detector can achieve a bit-error rate (BER) of 10-5 in the regions of low signal-to-noise ratio (SNR). Moreover, W-OTFS outperforms conventional OTFS in terms of peak-to-average power ratio (PAPR) and the complexity of modulation and demodulation processes.

Orthogonal time frequency space (OTFS) modulation and reconfigurable intelligent surface (RIS) are promising methods for next-generation wireless systems. We first incorporate the wavelet transform (WT) into the OTFS system. A simple matrix multiplication-based mathematical model is developed for the proposed wavelet-based OTFS (W-OTFS) system. Then, RIS-assisted communication is considered using the W-OTFS modulation. We use Haar and Daubechies wavelets for our investigations and observe that substantial improvements in terms of bit error rate (BER), peak-to-average power ratio (PAPR), and system complexity are attainable using the W-OTFS system. Our investigation reveals W-OTFS and RIS-assisted W-OTFS to exhibit superior performance compared to conventional OTFS in a high-mobility channel.

Orthogonal time frequency space (OTFS), a cuttingedge modulation technique, has gained attention as a potential prospect for future wireless communication. This paper incorporates wavelet transform in the conventional OTFS modulation, thereby presenting a method termed wavelet-aided OTFS (W-OTFS) modulation. The W-OTFS is then applied to optical wireless communication. In particular, a DC-biased optical W-OTFS (DCO W-OTFS) system is conceived, offering reduced spectral loss and peak-to-average power ratio (PAPR) in comparison to existing approaches. The proposed scheme outperforms both direct current optical (DCO) and asymmetrically clipped optical (ACO) OTFS schemes in terms of the PAPR and spectral efficiency. Additionally, the DCO W-OTFS is found to exhibit a lower bit error rate (BER) than the conventional DCO-based schemes.

Orthogonal time frequency space (OTFS) has emerged as a promising modulation for next-generation wireless systems. This two-dimensional (2D) modulation has the inherent capability of providing uninterrupted connectivity and improved performance in a high-mobility doubly-selective channel by mapping the information symbols to the delay-Doppler (DD) domain. In order to improve the directivity of OTFS signals, this paper considers a reconfigurable intelligent surface (RIS)-aided OTFS transmission for a high-speed environment. While OTFS mitigates the dispersion due to the highly mobile wireless channels by mapping the transmit symbols to the DD domain, RIS improves the directivity by appropriately shifting the phase of the signal in non-line-of-sight (NLOS) communication channels. We provide a concrete matrix-based mathematical model of the RIS-aided OTFS communication system. Capitalizing on the simple matrix multiplication-based model, a number of detectors can be used. These include the usual zero-forcing (ZF) and the minimum mean-squared error (MMSE) linear detectors, and a low-complexity message passing algorithm (MPA)-assisted detector. We evaluate the performance of these detectors for RIS-based OTFS systems. A deep learning (DL)-based signal detector is also proposed for the RIS-aided OTFS system. A significant improvement in bit-error-rate (BER) performance is achieved using the system considered, while the RIS-induced computational burden is not high. Furthermore, in NLOS communication, our system can ensure seamless connectivity with a reduced number of base stations (BS).

Fatal accidents are an inseparable part of life which often costs us loss of limbs especially hands and legs and turns any of our body asset into a burden to the family, as well as the society. The only solution to such misfortune is to facilitate the human with a new taste of living a happy life by having an artificial arm or leg that would be functional enough to have the everyday life activities smoother. The only way to make any artificial organ functional is to mimic its working from the remaining part of the organ or from the other similar organ more efficiently which is achieved by successfully and more intelligently extracting the features from biomedical signals especially from electromyogram (EMG) signals. This paper analyzes and detects various hand movements from surface EMG signals for six classes using a state-of-the-art Machine Learning scheme. This paper also presents an overview of the contemporary research on hand movement detection using EMG signals. The main contribution of this paper is the application of Convolutional Neural Network (CNN) in selection of the features from EMG signals along with some data preprocessing schemes including Fast Fourier Transform (FFT). The proposed approach has been applied to the available EMG dataset and demonstrates better accuracy along with computational simplicity than most existing schemes.

Pregnancy-associated anemia is a significant health issue that poses negative consequences for both the mother and the developing fetus. This study explores the triggering factors of anemia among pregnant females in India, utilizing data from the Demographic and Health Survey 2019-21. Chi-squared and gamma tests were conducted to find out the relationship between anemia and various socioeconomic and sociodemographic elements. Furthermore, ordinal logistic regression and multinomial logistic regression were used to gain deeper insight into the factors that affect anemia among pregnant women in India. According to these findings, anemia affects about 50% of pregnant women in India. Anemia is significantly associated with various factors such as geographical location, level of education, and wealth index. The results of our study indicate that enhancing education and socioeconomic status may serve as viable approaches for mitigating the prevalence of anemia disease developed in pregnant females in India. Employing both Ordinal and Multinominal logistic regression provides a more comprehensive understanding of the risk factors associated with anemia, enabling the development of targeted interventions to prevent and manage this health condition. This paper aims to enhance the efficacy of anemia prevention and management strategies for pregnant women in India by offering an in-depth understanding of the causative factors of anemia.

Right now, there is no perceptible vehicle management framework for Khulna university that provides a vehicle monitoring system, real-time fuel consumption data analysis, vehicle route optimization by calculating passenger traffic, and so on. Albeit creating that framework is the holistic objective of the total project, the significant objective of this paper is to propose a cost-effective, reliable IoT-based vehicle monitoring system for Khulna University based on the developed prototype device. The open-source controller and GPS-GPRS-based module determine the real-time position of the vehicle, and the location of the test vehicle can be communicated via GPRS, which is provided by the GSM network. This real-time location data is stored in a web-based IoT platform. Authorized users of the system can access this information via the internet. The proposed solution will facilitate all stakeholders, including teachers, students, and other Khulna University staffs, to use this information to make smarter travel selections. This will also pave the way for future research like intelligent vehicle route optimization by storing real data from different vehicles of Khulna University on the online database.

In recent years, the omnipresence of cardiac problems has been recognized as an epidemic. With the correct and quick diagnosis, both mortality and morbidity from cardiac disorders can be dramatically reduced. However, frequent medical check-ups are pricey and out of reach for a large number of people, particularly those living in low-income areas. In this paper, certain time-honored statistical techniques are used to determine the factors that lead to heart disease. Also, the findings were validated using various promising machine learning tools. Feature importance approach was employed to rank the clinical parameters of the patients based on the correlation of heart disease. In the case of statistical investigations, nonparametric tests such as the Mann Whitney U test and the Chi square test, as well as correlation analysis with Pearson correlation and Spearman Correlation were used. For additional validation, seven of the potential feature important based ML algorithms were applied. Moreover, Borda count was implemented to acknowledge the combined observation of those ML models. On top of that, SHAP value was calculated as a feature importance technique and for detailed evaluation. This research reveals two aspects of heart disease diagnosis.We found that eight clinical traits are sufficient to diagnose cardiac disorders, in which three traits are the most important sign of heart disease. One of the discoveries of this investigation uncovered chest pain, number of major blood vessels, thalassemia, age, maximum heart rate, cholesterol, oldpeak, and sex as sufficient clinical signs of individuals for the diagnosis of cardiac disorders. Over the above, considering the findings of all three approaches, chest pain, the number of major blood vessels, and thalassemia were identified as the prime factors of heart disease. The research also found, fasting blood sugar does not have a direct impact on cardiac disease. These findings will have the potency to be incredibly useful in clinical investigations as well as risk assessment for patients. Limiting the most critical features can have a significant impact on the diagnosis of heart disease and reduce the severity of health risks and death of patients.

Monitoring systems for electrical appliances have gained massive popularity nowadays.These frameworks can provide consumers with helpful information for energy consumption.Non-intrusive load monitoring (NILM) is the most common method for monitoring a household's energy profile.This research presents an optimized approach for identifying load needs and improving the identification of NILM occupancy surveillance.Our study suggested implementing a dimensionality reduction algorithm, popularly known as genetic algorithm (GA) along with XGBoost, for optimized occupancy monitoring.This exclusive model can masterly anticipate the usage of appliances with a significantly reduced number of voltage-current characteristics.The proposed NILM approach pre-processed the collected data and validated the anticipation performance by comparing the outcomes with the raw dataset's performance metrics.While reducing dimensionality from 480 to 238 features, our GA-based NILM approach accomplished the same performance score in terms of accuracy (73%), recall (81%), ROC-AUC Score (0.81), and PR-AUC Score (0.81) like the original dataset.This study demonstrates that introducing GA in NILM techniques can contribute remarkably to reduce computational complexity without compromising performance.

Heart failure is a chronic, irreversible condition. It causes 32% of entire mortality globally. The purpose of this study is to figure out which of the characteristics of patients with heart failure impacts their survival most so that improving those traits can lead to their quality health. Machine learning can be more proficient in diagnosing it precisely and minutely. XGBoost classified the data and genetic algorithm was further introduced for feature selection. Three features have been estimated to be most significant in our study-creatinine phosphokinase, serum sodium, and sex. This unearthing can aid doctors and physicians to determine the treatment they will be providing to the sufferers. Controlling the limits of creatinine phosphokinase and serum sodium can be impactful to reduce the severity of health hazards and mortality.

Human Activity Recognition (HAR) is one of the underlying research areas in the field of biomedical data science. One of the major tasks of researching with HAR datasets is extracting features from the raw datasets. Many methods have been proposed so far as an approach. In this paper, we have used two unorthodox methods (in the HAR field) for feature extracting, namely- 'Vector Point' (VP) and' Absolute Distance' (AD). By using these techniques, the data size gets decreased 3 and 6 times consecutively and the time complexity also gets decreased considerably. The extracted features are utilized as inputs to a Random Forest classifier. We have got an AVC score of '1' for our merged dataset with both the VP and AD methods which manifests the success of our work.

Activity recognition from human action data is quite a challenging task in the biomedical data science community. The main challenge in dealing with human activity recognition (HAR) datasets is their high cardinality. Therefore, reducing cardinality is a cardinal area of research in the HAR field. In this research, reducing the data dimensionality by utilizing future selection methods has been used. This research work has extracted features using wavelet packet transform (WPT) and the cardinality of the feature set has been reduced by using the Genetic Algorithm (GA) technique. The selected features also have been ranked according to their importance based on their SHAP values. In the venture, an interesting inspection has been found. That is in HAR datasets, signal values lay into lower frequency regions mostly. The highest accuracy and f1-score which have been got are 94.74%, 94.73%, and 89.98%, 89.67% for the feature extracted and feature selected dataset respectively.