Bannari Amman Institute of Technology

In the domain of passive brain-computer interface applications, the identification of emotions is both essential and formidable. Significant research has recently been undertaken on emotion identification with electroencephalogram (EEG) data. The aim of this project is to develop a system that can analyse an individual’s EEG and differentiate among positive, neutral, and negative emotional states. The suggested methodology use Independent Component Analysis (ICA) to remove artefacts from Electromyogram (EMG) and Electrooculogram (EOG) in EEG channel recordings. Filtering techniques are employed to improve the quality of EEG data by segmenting it into alpha, beta, gamma, and theta frequency bands. Feature extraction is performed with a hybrid meta-heuristic optimisation technique, such as ABC-GWO. The Hybrid Artificial Bee Colony and Grey Wolf Optimiser are employed to extract optimised features from the selected dataset. Finally, comprehensive evaluations are conducted utilising DEAP and SEED, two publically accessible datasets. The CNN model attains an accuracy of approximately 97% on the SEED dataset and 98% on the DEAP dataset. The hybrid CNN-ABC-GWO model achieves an accuracy of approximately 99% on both datasets, with ABC-GWO employed for hyperparameter tuning and classification. The proposed model demonstrates an accuracy of around 99% on the SEED dataset and 100% on the DEAP dataset. The experimental findings are contrasted utilising a singular technique, a widely employed hybrid learning method, or the cutting-edge method; the proposed method enhances recognition performance.

Wireless Sensor Networks present a significant issue for data routing because of the potential use of obtaining data from far locations with greater energy efficiency. Networks have become essential to modern concepts of the Internet of Things. The primary foundation for supporting diverse service-centric applications has continued to be the sensor node activity of both sensing phenomena in their local environs and relaying their results to centralized Base Stations. Malware detection and inadequate Cluster Heads node selection are issues with the current technology, resulting in a drastic decrease in the total Internet of Things-based performance of sensor networks. The paper proposes an Enhanced Lion Swarm Optimization (ELSO) and Elliptic Curve Cryptography (ECC) scheme for secure cluster head selection and malware detection in IoT-based Wireless Sensor Networks (WSNs). The paper includes network models, choice of Cluster Head (CH) and attack detection procedures. The proposed method chooses the Cluster Head with the best fitness function values, increasing data transmission speeds and energy efficiencies. Minimum Hop Detection has been implemented to provide the best routing paths against attack nodes. Security level for quick data transmissions via the Internet of Things using Wireless Sensor Networks strengthen sinkhole attacks and black hole nodes, which are successfully removed using this method. The proposed method integrates the use of Lion Swarm Optimization and Elliptic Curve Cryptography (ECC) enhances network security by ensuring secure data transmission and preventing unauthorized access, which is particularly important in IoT-WSN environments. The proposed method achieves less End delay, increased throughput of 93%, lower energy utilization of 4%, increased network lifetime of up to 96%, Packet Delivery Ratio of up to 98% and 97% of malicious node detection efficiently compared to existing methods.

All web applications remember the ultimate goal of storage services, Distributed Denial of Service (DDoS), to achieve high security from various attacks. The client-server application reduces fees and runs the elite registration gadget, while the notable part assumes that it is manufactured using a grant application. Analysis of the traceback data is difficult in the previous system, and then it containsa common DDoS attack dataset.The DDoS attacks involving different network layers such as (SIDDoS (SQL (Structured Query Language) injection), HTTP (HyperText Transfer Protocol) flood, TCP (Transmission Control Protocol) are recommended here as there are signs that new datasets of common datasets are not being collected in the proposed system. The DDoS attack classification consists of three main steps: preprocessing, trace out the source IP address, Backpropagation Shuffled Leaping Neural Network (BSLNN) based on maintaining the specifications’ integrity. The preprocessing is the first step based on the Gabor Filter used to remove the data’s noise and the second Adaboost Random Optimal Selection method to analyzethe packet flow’s relative frequency. The third step is the Backpropagation Shuffled Leaping Neural Network (BSLNN) classification of traceback data maintaining the specifications analysis’s integrity. The proposed system achieves good periodic study of packet flow accuracy and time complexity.

In recent years, the research on abnormal events detection is a significant work in surveillance video. Many researchers have been attracted by this work for the past two decades. As a result, several abnormal event detection approaches have been developed. Though several approaches have been used in the field still many problems remain to get the abnormal events detection accuracy. Moreover, many feature representations have limited capability to describe the content since several research works applied hand craft features, this type of feature can work in limited problems. To overcome this problem, this paper introduced the novel feature descriptor namely STS-D (Spatial and Temporal Saliency - Descriptor), which includes spatial and temporal information of the objects. This feature descriptor efficiently describes the shape and speed of the object. To find the anomaly score, fuzzy representation is modeled to efficiently differentiate the normal and abnormal events using fuzzy membership degree. The benchmark datasets UMN, UCSD Ped1 and Ped2 and real time roadway surveillance dataset are used to evaluate the performance of the proposed approach. Also, several existing abnormal events detection approaches are used to compare with the proposed method to evaluate the effectiveness of the proposed work.

This investigation presents an advanced biosensing platform integrating a Methylammonium lead halide metasurfaces with a dual-resonator architecture for high-precision waterborne pathogen detection. The engineered design incorporates a square ring resonator coupled with a graphene-functionalized circular resonator, optimized through systematic parametric analysis to achieve maximum sensitivity and specificity. Comprehensive electromagnetic simulations were performed utilizing COMSOL Multiphysics to characterize the sensor's electromagnetic response across multiple parameters, including graphene chemical potential modulation, incident wave angle variation, and resonator geometric configurations. The platform exhibited exceptional sensitivity to bacterial concentration-induced refractive index variations, demonstrating quantitative performance metrics of 488 GHzRIU⁻¹ sensitivity, 0.234 RIU detection limit, and a quality factor of 12. 914. Implementation of the XGBoost machine learning algorithm for sensor response optimization yielded optimal prediction accuracy (R² = 1.000) across all investigated parameters within the terahertz regime. These quantitative findings demonstrate the potential for integration of this sensing platform into high-throughput water quality monitoring systems, with significant implications for environmental surveillance and public health infrastructure applications.

The effective utilization and high‐value bioproducts from agro‐wastes make sense for a sustainable circular economy for agriculture. The article discusses the promising potential of utilizing agro‐wastes to produce high‐value bioproducts, particularly focusing on carbon dots (C‐dots) derived from such wastes. These C‐dots exhibit remarkable fluorescence properties and excellent biocompatibility, making them valuable nanomaterials for various applications. The dual sources of these C‐dots: green precursors sourced from both edible and non‐edible plant‐based materials, and chemical precursors involving acid and non‐acid reagents are highlighted. This diversity in precursor materials underscores the versatility and sustainability of C‐dot production. Importantly, the synthesis of fluorescent C‐dots achieved quickly and directly via hydrothermal carbonization, microwave technique, thermal pyrolysis carbonization, solvothermal technique, and ultrasonic process are review concisely intended for widespread application in fields ranging from bio‐imaging to optoelectronic devices. Furthermore, the article discusses the challenges associated with synthesizing high‐quality C‐dots from agro‐residues, indicating ongoing research efforts in this area. Likewise, key energy specific characteristics like optical, photoluminestic, photosimulated electron transfer, catalytic, mechanical, and carcinogenic attributes are discussed. Despite these energy specific characteristics, various energy applications of C‐dots, including their potential use in light‐emitting diodes, supercapacitors, and photovoltaics are outlined. This highlights the multifaceted nature of C‐dots and their contribution to advancing sustainable practices in agriculture while simultaneously addressing energy needs in various sectors. Overall, the article underscores the importance of leveraging agro‐wastes for the development of innovative and environmentally friendly bioproducts, contributing to the circular economy in agriculture.

Advancements in digital imaging and video processing are often challenged by low-light environments, leading to degraded visual quality. This affects critical sectors such as medical imaging, aerospace, and underwater exploration, where uneven lighting can compromise safety and clarity. To enhance image quality in low-light conditions using a computationally efficient system. This paper introduces an FPGA-based system utilizing the Retinex algorithm for low-light image enhancement, implemented on a Coarse-Grained Reconfigurable Architecture (CGRA). The system is designed using Verilog HDL on a Xilinx FPGA, prioritizing hardware optimization to achieve high-quality outputs with minimal latency. The system achieves a processing rate of 60 frames per second (fps) for images with a resolution of 720 × 576. Quantitative evaluations show a Peak Signal-to-Noise Ratio (PSNR) improvement to 43.18 dB, a Structural Similarity Index (SSIM) of 0.92, and a Mean Squared Error (MSE) reduction, demonstrating significant enhancements in image quality. The design also achieves a low power consumption of 0.186 W and efficient resource utilization, with only 2.2% of Slice LUTs and Slice Registers used. The FPGA-based system demonstrates significant improvements in image quality with high computational efficiency, proving beneficial for critical applications in various sectors.

In this study, an investigation has been performed on the glass fiber-vinyl ester composites with varying contents of silane treated white fumed silica at different thermal aging conditions. The main objective of this research work is to explicate the effectiveness of silane treated functional fumed silica on commercially used glass fibre-vinyl ester composite at elevated temperature aging conditions. The fumed silica was treated with 3-Aminopropyltrimethoxysilane (APTMS) and the composites were made by a hand layup principle. The composites were further subjected to temperature aging such as 50 & 60 °C @ 180 days and tested for their mechanical, flammability, and water absorption properties. The results indicate that, despite thermal aging the mechanical properties of the composites is stable, revealing the enhanced silane protection effect on glass fiber and fumed silica. Composite GST21, GST22, and GST23, subjected to thermal aging at 60 °C for 180 days, the tensile and flexural strengths experienced reductions compared to the un-aged specimens. However, the decrease in values is minimal, indicating the resilience of the composites even under harsh aging conditions. Flammability tests revealed consistent UL 94 ratings for all specimens, even after thermal aging. Water absorption percentages marginally increased after thermal aging, particularly at higher temperatures, yet remained within acceptable limits. SEM analysis provided visual evidence of the composite's microstructure, highlighting areas of voids, crack formations and low fiber pull-out. Despite some indications of degradation, the overall integrity of the composites remained sound, emphasizing the effectiveness of silane treatment on glass fiber and fumed silica in maintaining structural stability.In summary, while some minor losses in mechanical and physical properties were observed in the thermally aged specimens, the silane treated glass fiber and fumed silica mitigated these effects, ensuring that the composites retained their essential characteristics. Overall, the study underscores the importance of silane barrier protection on the reinforcement materials in enhancing the durability and performance of composite materials, even under challenging conditions.

The upsurge in urbanization is depleting natural resources due to the wide usage of aggregates in construction, posing a menace to further progress. If the contemporary state persists, it becomes a threat for further progress. As an initiative to preserve the natural resources, recycled aggregate is employed as a partial additive of fine aggregate, this practice endorses sustainability but demands a cautious investigation of its effect on concrete strength. Predicting the strength characteristics of concrete becomes crucial in this situation. Machine Learning (ML) algorithms estimate these properties, enabling engineers to evaluate and enhance the concrete mix with recycled materials, confirming durability and performance without solely depending on outdated testing methods The proposed work attempts how ML algorithms have been utilized to forecast the strength properties of concrete. Input samples were collected from the experimental setup for three grades of concrete (M15, M20, M25) with four various proportions (5%, 10%, 15% and 20%) at 7, 14, 21 and 28 ages of curing excluding the conventional mix. The ML models are developed by training the collected laboratory samples with respect to four parameters– cement, fine aggregate (FA), coarse aggregate (CA) and recycled aggregate (RA). The trained models predict the three properties of concrete– compressive strength, split tensile strength and flexural strength. The test outcomes are further validated using statistical measures such as correlation coefficient– R², Mean Square Error (MSE) and Mean Absolute Error (MAE) to assess the performance of the model. The performance evaluation of the proposed approach shows that The Support Vector Regressor outperforms the other ML models, achieving an R² value of about 90% and showing a reduced error rate in terms of MSE and MAE.

The tribological behavior of graphite (glass fiber) and silicon carbide (SiC) (glass fiber) composites is the main topic of this paper’s examination of recent and significant research trends in composite materials. The study analyzes wear characteristics with different applied loads in both dry sliding and oil lubrication conditions using a pin-on-disc the layout. The investigation shows that the load and velocity of the samples have an impact on wear rates and friction coefficients. According to experimental findings, oil-lubricated sliding has a lower wear rate and coefficient of friction than dry sliding. The study also reveals that, as compared to fiberglass without fillers, the use of graphite fillers in fiberglass composites dramatically reduces wear. The purpose of this study is to offer useful data for subsequent research on hybrid composite material wear analysis.

The spatio-temporal variations in stream water quality are primarily due to the heat and mass exchanges occurring at the interfaces which are propagative and dilutive along the stretch. The streams are also susceptible for increased pollutant loading from various point and non-point sources causing an imminent depletion of dissolved oxygen (DO) concentration. Though water quality parameters are several, it is pertinent to identify the most critical factors for an easy, reliable and meaningful monitoring system. It is also necessary to have a simplified data-driven model for making necessary predictions for various decision-making processes. The present study investigates the characteristic relationship between various forms of oxygen demands (carbonaceous biochemical oxygen demand—cBOD and sediment oxygen demand—SOD) in a stream with the available DO for a selected stretch of River Bhavani in Tamil Nadu, India. The modelling framework consists of simulating a data-extensive water quality profile using QUAL2K and integrating the outputs for minimizing the required count of parameters using a multi-perceptron feed-forward artificial neural network (ANN) model. The proposed methodology suggested that a minimum of five parameters (inorganic suspended solids—ISS, DO, cBOD, SOD and total nitrogen—TN) are sufficient to simulate the concentration profiles with reasonable accuracy (R2 varies from 0.88 to 0.95). For any local addition of organic-rich influent to the sediment load of the river, there is a corresponding change in the cBOD and DO within the selected stream profile. The results from the present study indicate the advantage of the combined modelling approach for prediction of water quality in case of complex interactions between various oxygen-demanding substrates.

Friction stir welding (FSW) is a solid-state joining method extensively utilized for dissimilar materials owing to its capacity to reduce imperfections and uphold material characteristics. Within this investigation, AA6061-T6 aluminum alloy and AZ31B magnesium alloy were subjected to friction stir welding under various process parameters, with the primary objective being the examination of the mechanical attributes of the welds and the juxtaposition of the forecasting precision of artificial neural networks (ANN) and response surface methodology (RSM) models. Key performance indicators such as tensile strength, microhardness, and impact energy were assessed. The obtained experimental data unveiled a spectrum of mechanical characteristics, encompassing tensile strength ranging from 69 to 144 MPa, microhardness from 79 to 112 HV, and impact energy from 7.9 to 8.55 J, with the validation test yielding a satisfactory outcome within an acceptable margin of error. The comparative scrutiny of ANN and RSM prognostications indicated that ANN achieved marginally lower average error percentages across all parameters, highlighting its enhanced predictive capacity. These observations emphasize the efficacy of both modeling methodologies in capturing the intricate interrelations inherent in dissimilar FSW procedures, with ANN demonstrating a slight edge in predictive precision.

Binary hydrogen bond liquid crystal (HBLC) complex is isolated from the combination of cholesteryl stearate (CHS) and 4-methoxycinnamic acid (4MCA) mesogenic compounds. Induced chiral smectic phases are observed and characterized using a polarizing optical microscope (POM). Enthalpy values and transition temperatures of liquid crystal (LC) phases and their thermal stability are reported using differential scanning calorimetry (DSC), and the same has been supported by the density functional theory (DFT) method. Observation of a strong peak at 3580 cm−1 from the FTIR spectrum confirms the hydrogen bond (H-bond) between chiral (CHS) and non-chiral (4MCA) mesogenic compounds. Further, DFT calculation explores the molecular mechanism of the complexation (CHS + 4MCA). The calculated global reactivity parameters (GRP) values (DFT calculations) justify the stability of the synthesized binary HBLC complex (CHS + 4MCA). In addition, topological analysis of DFT confirms the intermolecular interaction in the title binary HBLC complex. Donor and acceptor interaction in the binary HBLC complex (CHS + 4MCA) is reported using molecular electrostatic potential (MEP) analysis. Further, the impact of increased H-bond length and bond angle on induced chiral smectic ordering has been discussed in the present communication. Quantum theory of atoms-in-molecules (QTAIM) study endorses the existence of linear H-bonding of O–H···O in the binary HBLC complex. In addition, the enhanced mesomorphic behaviour along with non-linear optical properties (NLO) is reported.

Early detection of cervical cancer, a leading cause of morbidity and mortality among women globally, is crucial for successful treatment. Existing diagnostic methods, while effective, face limitations including cost, accessibility, and reliance on specialized personnel, especially in resource-constrained settings. Recent advancements in nanomaterials and metasurfaces offer promising avenues for developing highly sensitive, label-free biosensors capable of detecting cancer biomarkers at minute concentrations. This study presents the design and theoretical modelling of a novel metasurface-based sensor for cervical cancer detection, utilizing graphene, black phosphorus, and titanium dioxide as core sensing materials. The sensor exhibits dual-band operation (1.369–1.383 THz and 0.313–0.317 THz) with exceptional performance metrics: sensitivity reaching 400 GHzRIU⁻¹, a figure of merit of 5.882 RIU⁻¹, and quality factors ranging from 9.206 to 9.950. Furthermore,the sensor's dual-band functionality and 2-bit encoding capability suggest its potential for multi-parametric analysis and information processing, paving the way for more comprehensive diagnostic approaches. Additionally, integration of Support Vector Regression (SVR) with a Polynomial Kernel demonstrates remarkable performance, achieving an optimal R² score of 100%. This approach significantly reduces simulation time (80%) and resource requirements for sensor optimization.