Proposed Work Research Articles

New Results on Rapid Dynamical Pattern Recognition via Deterministic Learning From Sampling Sequences.

Rapid dynamical pattern recognition based on the deterministic learning method (DLM-based RDPR) aims to rapidly recognize the most similar dynamical pattern pair from perspectives of differences in inherent system dynamics. The basic mechanism is to use available recognition errors to reflect the differences in the dynamics of dynamical pattern pairs and then to make a decision based on a minimal recognition error (MRE) principle. This article focuses on providing a rigorous theoretical analysis of the MRE principle in DLM-based RDPR under the sampled-data framework. Specifically, we seek a unified methodology from the similarity definition to the measure implementation and then to derive general sufficient conditions and necessary conditions for the MRE principle. The main idea is to: 1) from the average signal energy aspect, define a time-dependent dynamics-based similarity in dynamical pattern pairs and reestablish the measure of recognition errors generated from the DLM-based RDPR; 2) introduce the energy-based Lyapunov method to establish the interrelation between the dynamical distance and the recognition error; and 3) derive sufficient conditions and necessary conditions from two directions of the interrelation. The proposed conditions distinguish themselves from virtually all of the existing DLM-based RDPR works with only sufficient conditions in the sense that it is shown in a rigorous analysis that under what conditions, the pattern pair recognized based on the MRE principle is indeed the most similar one. Therefore, the proposed work makes the DLM-based RDPR possess good interpretability and provides strong theoretical guidance in engineering applications.

HT-WSO: A hybrid meta-heuristic approach-aided multi-objective constraints for energy efficient routing in WBANs

Generally, Wireless Body Area Networks (WBANs) are regarded as the collection of small sensor devices that are effectively implanted or embedded into the human body. Moreover, the nodes included in the WBAN have large resource constraints. Hence, reliable and energy-efficient data transmission plays a significant role in the implementation and in constructing of most of the merging applications. Regarded to complicated channel environment, limited power supply, as well as varying link connectivity has made the construction of WBANs routing protocol become difficult. In order to provide the routing protocol in a high energy-efficient manner, a new approach is suggested using hybrid meta-heuristic development. Initially, all the sensor nodes in WBAN are considered for experimentation. In general, the WBAN is comprised of mobile nodes as well as fixed sensor nodes. Since the existing models are ineffective to achieve high energy efficiency, the new routing protocol is developed by proposing the Hybrid Tunicate-Whale Swarm Optimization (HT-WSO) algorithm. Subsequently, the proposed work considers the multiple constraints for deriving the objective function. The network efficiency is analyzed using the objective function that is formulated by distance, hop count, energy, path loss, and load and packet loss ratio. To attain the optimum value, the HT-WSO derived from Tunicate Swarm Algorithm (TSA) and Whale Optimization Algorithm (WOA) is employed. In the end, the ability of the working model is estimated by diverse parameters and compared with existing traditional approaches. The simulation outcome of the designed method achieves 13.3%, 23.5%, 25.7%, and 27.7% improved performance than DHOA, Jaya, TSA, and WOA. Thus, the results illustrate that the recommended protocol attains better energy efficiency over WBANs.

Novel Substitution Box Architectural Synthesis for Lightweight Block Ciphers

This paper proposes three novel Substitution box (S-box) architectures for lightweight block ciphers. The proposed S-box architectures are based on Boolean decomposition to achieve the design tradeoffs. Absorption of the Terminal Nodes property, positive Davio expansion, and Shannon decomposition are explored for the three S-box architectures. The proposed S-box variants are included in the SPN layer of the Midori-64 block cipher to validate their utility for a lightweight cipher. Also, to demonstrate the generality of the proposed logic decomposition approaches, the proposed work is expanded to the composite field architectures for the S-box of AES-128. Compared with state-of-the-art designs, the proposed designs exhibit a reduction of 66%, 36%, and 21% for energy, delay, and power metrics.

Robust And Discriminant Local Color Pattern (RADLCP): A novel color descriptor for face recognition

In [1] Karanwal et al. introduced the novel color descriptor in Face Recognition (FR) called as Fused Local Color Pattern (FLCP). In FLCP, the RGB color format is utilized for extracting features. From R, G and B channels, the MRELBP-NI, 6 × 6 MB-LBP and RD-LBP are imposed for feature extraction and then all are integrated to form the FLCP size. FLCP beats the accuracy of various methods. The one major shortcoming observed in [1] is that the basic format RGB is used for extracting features. Literature suggests that other hybrid formats achieves better recognition rates than RGB. Motivated from literature, the proposed work uses the hybrid color space format RCrQ for feature extraction. In this format R channel is taken from RGB, Cr channel is taken from YCbCr and Q channel is taken from YIQ. On R channel, MRELBP-NI is imposed for extracting features, On Cr channel 6 × 6 MB-LBP is imposed and on Q channel RD-LBP is imposed for extracting features. Then all channel features are joined to build the robust and discriminant feature called as Robust And Discriminant Local Color Pattern (RADLCP). Compression and matching is assisted from PCA and SVMs. For evaluating results GT face dataset is used. Results proves the potency of RADLCP in contrast to gray scale based implemented descriptors. RADLCP also beats the results of FLCP. Several literature techniques are also outclassed by RADLCP. For evaluating all the results MATLAB R2021a is used.

An Improvised Modulation and Control Approach for Dual Active Bridge DC–DC Converter System

Aimed at the control and modulation of dual active bridge bi-directional dc (DAB-BIDC) converter, an unconventional approach is applied to explore its effect in symmetrical wave-based isolated dc-dc converters. The existing control schemes of DAB-BIDC converter with unified or optimal phase shift (UPS/OPS) modulation have the constraints like complex control schemes, light load efficiency, current stress, and reactive power. Since a 50 Hz, grid-connected conventional converter system uses the sinusoidal pulse width modulation method to minimize the requirement for passive filters. Similarly, this work investigates the same analogy for the high-frequency (HF) symmetrical wave converters to get effective modulation and control scheme for the DAB-BIDC converter. Furthermore, the proposed sinusoidal pulse width-phase shift (SPW-PS) modulation scheme has been implemented by the optimally derived control parameters using Lagrange's Multiplier Method (LMM) to minimize the reactive power flow and current stress reduction at the HF link. Therefore, the DAB-BIDC converter with the proposed modulation and control approach can be used for EV charging and solid-state transformers applications. Finally, the proposed work is validated using MATLAB simulation and experimental setup having turn ratio 36/12 V, fundamental frequency 5-kHz, and switching frequency 100-kHz for the experimental validation using OPAL-RT.

Event-Based Background-Oriented Schlieren.

Schlieren imaging is an optical technique to observe the flow of transparent media, such as air or water, without any particle seeding. However, conventional frame-based techniques require both high spatial and temporal resolution cameras, which impose bright illumination and expensive computation limitations. Event cameras offer potential advantages (high dynamic range, high temporal resolution, and data efficiency) to overcome such limitations due to their bio-inspired sensing principle. This article presents a novel technique for perceiving air convection using events and frames by providing the first theoretical analysis that connects event data and schlieren. We formulate the problem as a variational optimization one combining the linearized event generation model with a physically-motivated parameterization that estimates the temporal derivative of the air density. The experiments with accurately aligned frame- and event camera data reveal that the proposed method enables event cameras to obtain on par results with existing frame-based optical flow techniques. Moreover, the proposed method works under dark conditions where frame-based schlieren fails, and also enables slow-motion analysis by leveraging the event camera's advantages. Our work pioneers and opens a new stack of event camera applications, as we publish the source code as well as the first schlieren dataset with high-quality frame and event data.

An optimal model for detection of lung cancer using convolutional neural network

In terms of frequency and mortality, lung cancer ranks second among all cancers worldwide for both men and women. It is suggested that pattern classification and machine learning be applied to the identification and categorization of lung cancer. Convolution neural network (CNN) techniques divide the input data into groups according to the distinctive characteristics of the input. Using a standard approach to analyze a large number of computed tomography images, early detection of lung cancer can save lives. The suggested effort is centered on identifying the precise type of cancer and making predictions about whether it is benign or aggressive. The deployment of proposed model is an attempt to improve the accuracy of the system. The proposed work showed an overall accuracy of 98.4% during the detection of lung cancer and 98.8% accuracy towards the prediction of specific type in the lung cancer. Mean average precision score of 97.17% and 98.75% test and validation respectively. 0.96, 0.93, and 0.95 for malignant test data.

Federated learning-based trajectory prediction for dynamic resource allocation in moving small cell networks

With the evolution of fifth generation (5G) of mobile communication, vehicular edge computing (VEC) and moving small cell (MSC) network are gaining attention because of their capability to provide improved quality-of-service (QoS) to vehicular users. The ultimate goal of VEC-enabled MSC network is to diminish the vehicular penetration effect and path loss resulting in improved network performance for vehicular users. In this paper, we explore distributed resource block (RB) allocation for fronthaul links of MSCs with probabilistic mobility in VEC-enabled MSC network. The proposed work exploits the computational power of road side units (RSUs) deployed with VEC servers present along the road sides to allocate the resources to MSCs in a distributed manner by exchanging limited information, with the objective of maximizing the data rate achieved by MSC network. Moreover, we propose federated learning (FL)-based position prediction of MSCs to predict the trajectory of MSCs in advance for efficient prediction of resource allocation in MSC network. Simulations results are presented to compare the position prediction dependent distributed and centralized time interval dependent interference graph-based resource allocation to MSCs in terms of RB utilization and average achievable data rate of MSC network. For comparison, we further investigated centralized as well as distributed threshold time dependent interference graph-based allocation of resources to MSC network.

Power quality investigation with multilevel inverter by photovoltaic-fed dynamic voltage restorer

ABSTRACT In this research endeavor, a novel approach has been presented to enhance the power quality through a Photovoltaic (PV)-based Dynamic Voltage Restorer (PV-DVR). The proposed work is designed to address voltage fluctuations, flicker, and harmonics within medium or low voltage micro-grid systems. Unlike existing DVR systems that often neglect injected harmonics and converter losses, in the proposed work there is an integration of a PV-fed multilevel inverter system employing Selected Harmonic Elimination and Moderation-based Pulse Width Modulation (SHEAM-PWM). This innovative configuration, coupled with a DC – DC boost converter and a series injection transformer, not only aims to alleviate voltage stress on devices but also optimizes the number of switches to minimize switching losses. The proposed setup successfully reduces the overall voltage Total Harmonic Distortion (THD) to a significant extent, delivering superior dynamic performance. To determine optimal switching angles for the SHEAM-PWM technique, ensuring the elimination or mitigation of the maximum number of harmonics from the output voltage waveform, we employ the Biogeography-Based Optimization (BBO) approach. The computation of these angles is performed offline to eliminate lower-order harmonics, and the results are stored in the microcontroller’s memory using mixed model equations for online applications. The efficacy of our proposed method is demonstrated through both simulation and experimental validation in a real-time laboratory system. The simulation and experimental results obtained underscore the potential utility of our technique for Dynamic Voltage Restorer (DVR) applications, particularly those requiring low harmonic injection. The proposed method introduces a comprehensive and innovative solution for power quality enhancement, combining PV technology, multilevel inverter systems, and advanced optimization techniques. The proposed PV-DVR system exhibits promising results, showcasing its potential for practical implementation in residential micro-grid systems with a focus on minimizing harmonics and ensuring efficient grid compensation.

Detecting Adversarial Examples Utilizing Pixel Value Diversity

In this article, we introduce two novel methods to detect adversarial examples utilizing pixel value diversity. First, we propose the concept of pixel value diversity (which reflects the spread of pixel values in an image) and two independent metrics (UPVR and RPVR) to assess the pixel value diversity separately. Then we propose two methods to detect adversarial examples based on the threshold method and Bayesian method respectively. Experimental results show that compared to an excellent prior method LID, our proposed methods achieve better performances in detecting adversarial examples. We also show the robustness of our proposed work against an adaptive attack method.

Multi-sensor fusion based optimized deep convolutional neural network for boxing punch activity recognition

Recent advancements in deep learning have significantly enhanced the recognition of player activities in sports by enabling automatic feature extraction. In our proposed work, we focus on recognizing six distinct punches in the context of boxing. We incorporate the sliding window technique during the pre-processing stage to transform the time-series data obtained from Inertial Measurement Unit (IMU) sensors in a boxing punch activity recognition system. Our approach influences a sensor fusion-based Deep Convolutional Neural Network (DCNN) classification model to identify various boxing punches accurately, achieving an impressive F1 score. The system demonstrates proficiency in distinguishing similar activities, such as jab and hook punches where the existing systems made misclassifications due to subtle variations in arm flexion that differentiate the two. Through experimentation, we identify an optimal window size for boxing punch activity recognition, which falls within the range of 15–20 frames (equivalent to 0.15–0.25 s). This window size selection results in a notable reduction in inference time. To evaluate our proposed model, we conduct comparisons with a standard DCNN and an optimized DCNN model. Our proposed optimized DCNN model demonstrates enhanced recognition accuracy, achieving an impressive 99%, coupled with an improved F1 score of 87%. Furthermore, the model displays a remarkable reduction in inference time, clocking in at less than 1 ms. Overall, our research contributes to the field of sports-related player activity recognition by employing the power of deep learning. By expertly combining these techniques, we achieve remarkable accuracy, precision, and efficiency in recognizing various boxing punches.

An Intelligent Optimized Control System for Induction Heating Application

In the modern era, induction heating (IH) technology is utilized for industrial functions like medical applications and domestic applications. Furthermore, efficient uses such as control topologies, magnetic component design, and power electronics have high demand as well as cost-effective with respect to advanced technology. However, IH technology directly targets the power converter and coil so it has wasted higher amount of heat. Also, high power dissipation in an IH system leads to high energy costs and reduced system performance. Hence, various approaches are developed to minimize power dissipation. However, they cannot reduce power dissipation in the induction coil. Thus, a novel Decision-based Squirrel Control System was developed in this article to decrease the power wastage in induction coil optimally. The developed model’s squirrel fitness function monitors power wastage and compensates them by tuning it. The developed model is executed in MATLAB/Simulink, and the results are determined. In terms of rising time, settling time, overshoot, and error rate. Here, the proposed work can achieve 0.075s resting time, 1.02s settling time, 0.05s overshoot, and 0.29% of error rate. The comparative assessment proves that the presented approach attained better results than others.

Simulation and FPGA based Implementation of PI and Fuzzy Controllers for Cascaded Asymmetric Multilevel Inverter

The main objective of this work is to improve the performance of Multilevel Inverter by using the novel LS-PD- PWM and controlling technique instead of the existing approach. The proposed work focus on closed loop controller such as existing PI and Fuzzy Logic Controller for resistive load with Multilevel Inverter. The proposed Asymmetric fifteen level Inverter offers less THD and high frequency than the Asymmetric nine level inverter. Asymmetric multilevel inverter is used to reduce the switching losses and THD. In this work is the generation and implementation of Pulse Width Modulation (PWM) techniques using Field Programmable Gate Array (FPGA) for Multilevel Inverter. PWM can be generated via Microprocessors, Microcontroller, FPGA and DSP Processors. Higher density Programmable Logic Devices (PLD) such as FPGAs can be used to amalgamate large amounts of logic into a single Integrated. Pulse width modulation (PWM) is generally used in power converter control. To generate LS-PD- PWM by using FPGA, this occupies less memory capacity and accurate than conventional one. An efficient PWM generation technique is performed using FPGA for lower frequency applications. Simulation and synthesis is done by ModelSim6.3c and Xilinx10.1. Further, the Implementation is carried out by FPGA Spartan3 Xc3s50. Key Word: PWM, FPGA, PLD (Programmable Logic Devices), LS-PD-PWM

NSDIE: Noise Suppressing Dark Image Enhancement Using Multiscale Retinex and Low-Rank Minimization

It is inevitable for dark images to have crucial information obscured by low-light conditions, which are worsened by the presence of noise in these images. This work introduces a groundbreaking solution, Noise-Suppressing Dark Image Enhancement for Web Apps (NSDIE), to address the challenging task of enhancing low-light images marred by noise. The proposed work utilizes a low-rank model with simultaneous enhancement of reflectance and illumination components to improve the nighttime scenes while also eradicating the present noise of the image. The reflectance component is further processed using a multiscale retinex model to compensate for the possible color distortions while the illumination component is enhanced using the camera response model to ensure the genuineness of the scene. The proposed work is also tested for a standalone application and is presented to the user through a web portal to aid the concerns of dark image enhancement in the daily life of the user. Rigorous quantitative and qualitative analyses assert NSDIE's superiority over existing techniques, establishing its pivotal role in addressing the critical concern of dark image enhancement.

Adaptive Data Transmission Protocols for Energy Harvesting WSNs Used in Agriculture

Energy consumption is a major concern in wireless sensor networks (WSNs) as it affects the lifespan of sensor nodes. Battery-based WSNs have a short operating period, which makes them impractical for real-time applications, for instance in agriculture. Energy harvesting and suitable medium access control (MAC) protocols have been used to extend the lifetime of nodes. Receiver-initiated protocols have been proved to be the best solution for energy harvesting WSNs. However, they suffer from a key disadvantage, i.e. an increase in collision rate. These collisions need to be reduced using a multi-layer protocol structure. In such a context, a new solar-based hybrid MAC (SHMAC) protocol relying on receiver-initiation and characterized by a multi-layer structure is proposed. It is an adaptive protocol capable of adapting to changing weather conditions. The nodes with a high energy harvesting rate have a higher level of residual energy and are active for longer time periods compared with those with low energy harvesting characteristics. The proposed work has shown improvements in two major MAC layer parameters, i.e. collision rate and energy neutrality operation ratio (ENO).

A novel UEBASC‐qZSI‐based BLDC motor drive with modified DTC for EV application

SummaryAt present taking into account the situation of the environment and the growing demand of fossil fuels, electric vehicles (EVs) are a highly promising option for the transportation system. Many manufacturers of electric vehicles employ BLDC motors as the vehicle's driving component. In contrast to the many advantages that BLDC motors offer, one of the major drawbacks of employing these motors is the production of relatively high distortion in electromagnetic torque that is created by commutation. This distortion in torque has an effect on the vehicle's performance. In this proposed work a BLDC motor is powered by a cutting‐edge, highly gained q‐ZSI Ultra‐Extended Boost Active Switch Controlled Based Quasi Z‐source Inverter, or UEBASC‐qZSI. In the aforementioned drive, the torque distortion is minimized with the use of a modified direct torque control approach. With the aid of this UEBASC‐qZSI, the efficiency of EVs can be improved, and the overall safety of the driving cycle is also increased. Any vehicle's efficiency can be enhanced with the help of this UEBASC‐qZSI, and the overall safety of the overall system can also be improved. To validate the efficacy of the system, a Simulink model is developed in the MATLAB platform. Also, a hardware prototype model is developed for experimental authentication. The modeling and investigational findings demonstrate that the proposed system has significantly reduced the torque ripple while obtaining the necessary input dc bus voltage.

Meeting Power Challenges with TCSC and Next Generation TCSC: SPLIT TCSC

The electricity demand is continuously increasing due to population growth and development. There is a lack of generation and transmission facilities, so the existing system is used to transfer more power to meet the rising power demand. This causes overexploitation of the present power system and gives rise to stability problems. The power transfer capacity of the existing transmission lines can be increased by using fixed capacitors by varying the impedance of the transmission lines, but they have associated SSR problems. In this situation FACTS devices TCSC are used to control power flow in the system. TCSC mitigate power oscillation, stability problems and various power quality problems like voltage sag and voltage swell created due to disturbances or faults. In the proposed work the TCSC is used to increase the power flow capacity of the system without the need for investment in new lines and systems. The active power transfer with different controllers is discussed. The two modules of TCSC with different ratings are taken and analysed. Then there are some problems in ordinary TCSC too. On tuning the ordinary TCSC in the critical region of the reactance characteristic curve there is observed a large gap of reactance at both the inductive and capacitive critical regions. Because of the large elapse/gap of reactance in this region small change of power demand is not possible in the power system which reduces the flexibility of the power system, and it becomes rigid. These challenges of ordinary TCSC are met here by the next generation device of TCSC: Split TCSC in the power system. Split TCSC tunes the critical region reactance with many firing points and hence elapse of reactance is very small which allows for fine tuning of power flow over the transmission line thus mitigating the problems of ordinary TCSC. Split TCSC includes all the other benefits of ordinary TCSC such as increased power flow, stability improvement, damping oscillations, and mitigating SSR as it belongs to the family of TCSC. Thus, in the proposed work the power challenges are met using TCSC and Split TCSC.

Hybrid model for brain tumor detection using convolution neural networks

<div>The development of abnormal cells in the brain, some of which may turn out to be cancerous, is known as a brain tumor. Magnetic resonance imaging (MRI) is the most common technique for detecting brain tumors. Information about the abnormal tissue growth in the brain is visible from the MRI scans. In most research papers, machine learning (ML) and deep learning (DL) algorithms are applied to detect brain tumors. The radiologist can make speedy decisions because of this prediction. The proposed work creates a hybrid convolution neural networks (CNN) model and logistic regression (LR). The visual geometry group16 (VGG16) which was pre-trained model is used for the extraction of features. To reduce the complexity, we eliminated the last eight layers of VGG16. From this transformed model, the features are extracted in the form of a vector array. These features fed into different ML classifiers like support vector machine (SVM), and Naïve Bayes (NB), LR, extreme gradient boosting (XGBoost), AdaBoost, and random forest for training and testing. The performance of different classifiers is compared. The CNN-LR hybrid combination outperformed the remaining classifiers. The evaluation measures such as Recall, precision, F1-score, and accuracy of the proposed CNN-LR model are 94%, 94%, 94%, and 91% respectively.</div>

Dual axis solar tracking system for agriculture applications using machine learning

The three most basic amenities required for human survival are food, shelter and clothing. In today's tech-savvy generation, these have experienced a great deal of scientific advancement. Unfortunately, agriculture is still more man power-oriented. So they have to rely on the hit and trial method to learn from experience which leads to waste of time. In proposed work includes an automated system using dual axis solar tracking system and gives crop recommendation for different types of soil to yield maximum. The suggested system is a dual-axis solar tracker based on machine learning that is intended to considerably increase the effectiveness of energy harvesting. The approach makes use of the logistic regression algorithm (LR) to do this. This novel strategy tries to maximize the solar panel's ability to produce energy, leading to increased energy yields. The quality of soil is predicted by using suitable sensors for crop recommendation. The data’s are temperature, humidity, pH of soil, nitrogen, phosphorous & potassium in soil and rainfall in soil are considered. For crop recommendation six algorithms- SVM, KNN, Native Bays, Logistic Regression, Decision Tree classifier, Random Forest Classifier are applied and tested. It is found that random forest classifier gave us excellent results.

Hybrid model for detection of brain tumor using convolution neural networks

The development of aberrant brain cells, some of which may turn cancerous, is known as a brain tumor. Magnetic resonance imaging (MRI) scans are the most common technique for finding brain tumors. Information about the aberrant tissue growth in the brain is discernible from the MRI scans. In numerous research papers, machine learning, and deep learning algorithms are used to detect brain tumors. It takes extremely little time to forecast a brain tumor when these algorithms are applied to MRI pictures, and better accuracy makes it easier to treat patients. The radiologist can make speedy decisions because of this forecast. The proposed work creates a hybrid convolution neural networks (CNN) model using CNN for feature extraction and logistic regression (LR). The pre-trained model visual geometry group 16 (VGG16) is used for the extraction of features. To reduce the complexity and parameters to train we eliminated the last eight layers of VGG16. From this transformed model the features are extracted in the form of a vector array. These features fed into different machine learning classifiers like support vector machine (SVM), naïve bayes (NB), LR, extreme gradient boosting (XGBoost), AdaBoost, and random forest for training and testing. The performance of different classifiers is compared. The CNN-LR hybrid combination outperformed the remaining classifiers. The evaluation measures such as recall, precision, F1-score, and accuracy of the proposed CNN-LR model are 94%, 94%, 94%, and 91% respectively.