National Institute of Technology, Hamirpur

Federated learning holds significant potential as a collaborative machine learning technique, allowing multiple entities to work together on a collective model without the need to exchange data. However, due to the distribution of data across multiple devices, federated learning becomes susceptible to a range of attacks. This paper provides an extensive examination of the different forms of attacks that can target federated learning systems. The attacks discussed include data poisoning attacks, model poisoning attacks, backdoor attacks, Byzantine attacks, membership inference attacks, model inversion attacks, etc. Each attack is examined in detail, with examples from the literature provided. Additionally, potential countermeasures to defend against these attacks are explored. The objective of this review is to provide an in-depth survey of the current landscape in federated learning attacks and corresponding defense mechanisms.

Soil stabilization is crucial for enhancing the engineering properties of soil and constructing durable infrastructure, such as highways, airports, and roadways. The study's constituents were previously employed separately, and the soil's strength improved when they were coupled with other ingredients. Experimental investigations were conducted to assess the effects of varying proportions of C&D waste, CCR, and molasses on key soil characteristics, including compaction, shear strength, and plasticity. A series of crucial tests, including Atterberg limits, compaction characteristics, differential swell index, unconfined compressive strength (UCS), California Bearing Ratio (CBR), and Scanning Electron Microscope (SEM) analysis, were conducted to evaluate the performance of the stabilized soil. Test results indicated marked improvements in the Atterberg limits, reduced swell potential, and elevated values of UCS and CBR, demonstrating the effectiveness of the proposed stabilization method. CDW, CCR, and molasses enhance Unconfined Compressive Strength (UCS) by improving strength and cohesion. The addition of these chemicals significantly improved the performance of the soil, as seen by the decreased settling, enhanced strength, and greater infrastructure durability. Molasses served as an effective natural binder, while glass fibers improved tensile strength and durability by distributing stress evenly. This approach addresses waste management issues and promotes sustainable construction practices, offering a cost-effective solution for enhancing soil performance and paving the way for resilient infrastructure development.

This study incorporates the formation of carbon quantum dots (CQDs) via a hydrothermal approach, recording the first-time use of castor leaves as a natural precursor. The used precursor offers various benefits including novelty, abundance, elemental composition, and biocompatibility. CQDs were further characterized with multiple techniques including high-resolution transmission electron microscope (HR-TEM), X-ray photoelectron microscopy (XPS), X-ray diffraction (XRD), Fourier-transform infrared spectroscopy (FTIR), Raman spectroscopy, UV–visible spectroscopy, Zeta analysis, and optical spectroscopy. They are fundamentally composed of carbon (71.37%), nitrogen (3.91%), and oxygen (24.73%) and are nearly spherical, and uniformly distributed with an average diameter of 2.7 nm. They possess numerous interesting characteristics like broad excitation/emission bands, excitation-sensitive emission, marvelous photostability, reactivity, thermo-sensitivity, etc. A temperature sensor (thermal sensitivity of 0.58% C−1) with repeatability and reversibility of results is also demonstrated. Additionally, they were found selective and sensitive to ions in aqueous solutions. So, they are also utilized as a fluorescent probe for metal ion (Fe3+) sensing. The lowest limit of detection (LOD) value for the current metal ion sensor is 19.1 µM/L.

This paper presents a comprehensive review of finite element methods (FEMs) applied to a diverse range of reaction-diffusion equations (RDEs). Beginning with a historical overview of FEMs, we then provide a summary of various FEMs, including standard Galerkin (both conforming and non-conforming), mixed Galerkin, discontinuous Galerkin, and weak Galerkin. Additionally, a priori and a posteriori error have been discussed for standard Galerkin. In further discussion related to RDEs, we provide insights into the evolution of these equations and their significance in various fields. We then systematically review these FEMs for solving different types of RDEs, including more recent advances pertaining to RDEs with nonlinear reaction terms, and advection reaction-diffusion equations. Finally, we briefly highlight the applications of machine learning and deep neural networks to FEM.

Non‐orthogonal multiple access (NOMA), as a multiple access scheme, is foreseen as an emerging technology in enhancing the achievable sum rate of indoor downlink multiple‐input multiple‐output (MIMO) visible light communication (VLC) networks. Also, power allocation scheme plays a vital role in achieving these enhanced achievable sum rates. In literature, there are multiple power allocation schemes available such as gain ratio power allocation (GRPA), normalized gain difference power allocation (NGDPA), normalized logarithmic gain ratio power allocation (NLGRPA), and so forth. In this paper, an improved power allocation scheme has been proposed, which is based on the ratio of logarithmic sum channel gains of users as well as on the user index in the decoding order. In the proposed power allocation scheme, multiple transmitting light emitting diodes (LEDs) and multiple users are considered along with their optical channel information required for data communication. The transmitting LEDs are fixed on the room ceiling at optimal positions. The optimal positions of LEDs are selected by using particle swarm optimization (PSO) algorithm. The simulation results depict that the proposed power allocation scheme with optimal LED positions can achieve a performance gain of 5% or higher in the achievable sum rate over NLGRPA scheme when the number of users present in the room is more than five, and even when the user location is far from the LED. Thus, the proposed power allocation scheme with optimal LED positions achieves significant performance improvement when compared with existing power allocation schemes for NOMA‐based MIMO‐VLC networks for randomly located users in indoor scenario.

To achieve sustainability in the water treatment processes through electrocoagulation, optimal operating parameters are required for its proper functioning and to fulfil environmental goals. Further, optimal values for electrocoagulation depend on accurate physical and numerical models. These models simulate the complexities of the treatment process. This study explores the capabilities of the deep learning modelling tool artificial neural network (ANN) for multi-objective modelling. ANN models the removal of arsenic and fluoride from the water with respect to current density, pH, time, and initial concentrations of arsenic (As) and fluoride (F). This study indicates that ANN models have higher accuracy than isotherm models for representing the electrocoagulation process as reported higher coefficient of determination values of ANN (As:0.9995, F:0.9990) over isotherm models (Langmuir, Freundlich and Sips Models). To achieve the sustainability and efficiency of the electrocoagulation process, the best model, in this case, the ANN model, used to find optimal values whereby effective remediation of the arsenic and fluoride from the water can be achieved. For the same, multiple metaheuristic optimisation tools are explored on the objective function created from the ANN model. This study explored genetic algorithm (GA), particle swarm optimisation (PSO), and ant colony optimisation (ACO) meta-heuristic tools and compared these tools on economic terms. Upon validation of these optimal conditions, it was reported that PSO (2.62 mAcm⁻², pH: 6.5, 45 min, F: 10mgL⁻¹ and As: 450 µgL⁻¹) showed more correlated and economical operating conditions (1.584USDm⁻³) to achieve the 96% fluoride removal and 97% arsenic removal. Graphical Abstract

The Internet of Things (IoT) enables 5G communication, and it is widely used in many industries owing to its ability to connect a massive number of devices, process data intelligently, and enable remote control. However, the open nature of wireless communications makes information security and privacy critical concerns. In this paper, we investigate the issue of physical layer security (PLS) within a two-hop wireless cooperative network, where communication is facilitated by an untrusted relay. This investigation was performed over mixed Rayleigh-Nakagami-$\boldsymbol{m}$ fading channels in the presence of multiple non-colluding eavesdroppers. To improve the system’s security, we considered the incorporation of cooperative jamming and opportunistic relaying techniques. We formulated explicit equations for both the secrecy capacity and the secrecy outage probability within the system. Finally, by leveraging numerical results, simulations, and comparisons with existing works, we demonstrated the effectiveness of the proposed approach. This demonstration highlights that the suggested method serves as an effective method to enhance the security of IoT data collection processes, particularly in scenarios involving untrustworthy relays and highly risky eavesdropping threats.

This study investigated the impact of antioxidants diphenylamine (DPA) and nanoparticles ceria (CeO2) on engine block vibration in a B30 biodiesel blend. The influence of critical input parameters such as compression ratio (CR), fuel injection pressure (FIP), load, and exhaust gas recirculation (EGR-HOT) on vibration behavior was analyzed. Response Surface Methodology (RSM) and Machine Learning (ML) algorithms were employed to predict experimental root mean square (RMS) acceleration values. Experiments were conducted using diesel and biodiesel blends (B30, B30+DPA100, and B30+DPA50+CeO250) under varying conditions of CR, FIP, load, and EGR-HOT. ML algorithms, including Support Vector Machines (SVM), Multi-Layer Perceptron (MLP), and k-Nearest Neighbors (k-NN), were employed to predict RMS acceleration. Model performance was evaluated using metrics such as coefficient of determination (R2), Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and relative RMSE (rRMSE). The study finds significant reductions in mean RMS acceleration for B30, B30+DPA100, and B30+DPA50+CeO250 compared to diesel across load, CR, FIP, and EGR-HOT variations, demonstrating the effectiveness of these fuel blends in minimizing engine vibration. For load variation, the reductions were 12.32% for B30, 7.15% for B30+DPA100, and 9.12% for B30+DPA50+CeO250 as compared to diesel. Similarly, with CR variation, the decreases were 12.14% for B30, 7.04% for B30+DPA100, and 8.99% for B30+DPA50+CeO250 as compared to diesel. For FIP variation, the mean RMS acceleration reductions were 10.28% for B30, 5.97% for B30+DPA100, and 7.61% for B30+DPA50+CeO250 as compared to diesel. Finally, with EGR-HOT, the mean RMS acceleration reductions were 10.33% for B30, 5.99% for B30+DPA100, and 7.65% for B30+DPA50+CeO250 as compared to diesel. Furthermore, findings show that the k-NN model outperforms SVM and MLP models, exhibiting superior R2 values and consistently lower error metrics. Overall, these findings emphasize the efficacy of DPA and CeO2 additives in mitigating engine vibrations in biodiesel blends and k-NN algorithm emerged as a reliable tool for predicting RMS acceleration.

Metal matrix composites (MMCs) have attracted the interest of researchers due to its potential to improve tribological and mechanical properties. This study investigates the influence of ceramic reinforcement of TiC/CaF2 on physical, mechanical and friction characteristics of CuNi alloy/TiC MMCs. The composites reinforced with CaF2 particles (0, 4, 8 wt pct) were developed by powder metallurgy, using solid-state sintering. The homogeneous dispersion of TiC/CaF2 particles in the Cu–Ni matrix was confirmed by microstructure analysis. The CuNi/10 wt pct TiC composite exhibited the highest density (7.32 g/cm3) and hardness (40.6 HV/0.1). Among all the composites tested, the lowest COF of (0.21) was observed for CuNi + (10 wt pct)TiC + (8 wt pct)CaF2. The COF of the CuNi + (10 wt pct)TiC + (8 wt pct)CaF2 composite is reduced by 62.5 pct compared to the current material. Friction-induced chemical reactions resulted in the development of oxides (Cu2O, NiO, Fe2O3, and TiO2 phases) at the contact interface, substantially influencing tribological characteristics and wear mechanisms. Delamination, tribo-oxidation wear, adhesive wear, and abrasive wear were the main wear mechanisms observed in the fabricated composites.

This study investigates the impact of local thermal non-equilibrium on the stability analysis of partially ionized plasma within a porous medium. The plasma, heated from below, is enclosed by various combinations of bounding surfaces. Both nonlinear (via the energy method) and linear (utilizing the normal mode analysis method) analyses are performed. Eigenvalue problems for both analyses are formulated and solved using the Galerkin method. The study also explores the effects of compressibility, medium permeability and magnetic fields on system stability. The collisional frequency among plasma components and the thermal diffusivity ratio significantly influence energy decay. The results reveal that the Rayleigh–Darcy number is identical for both nonlinear and linear analyses, thus eliminating the possibility of a subcritical region and confirming global stability. The principle of exchange of stabilities is validated, indicating the absence of oscillatory convection modes. Medium permeability, heat-transfer coefficient and compressibility delay the onset of convection, demonstrating stabilizing effects. Conversely, the porosity-modified conductivity ratio hastens the convection process, indicating destabilizing effects. Rigid–rigid bounding surfaces are found to be thermally more stable for confining the partially ionized plasma. Additionally, the magnetic field exerts a stabilizing influence.

The modal analysis identifies a structure's inherent dynamic characteristics, such as eigenfrequency and vibrational mode shapes. It has been widely used for many different purposes during the last few decades. Still, more consideration must be given to investigating the modal characteristics of the multi-paneled concrete pavement airport runways. This study offers a finite element (FE) based modeling approach for the modal analysis of a concrete pavement airport runway, specifically focusing on a jointed plain concrete pavement (JPCP) system, which will aid in improving runway pavement design, precisely adjusting the eigenfrequencies by distributing mass and stiffness of the structure to avoid resonance and ensure structural integrity. A FE model in three dimensions of a concrete pavement runway was developed and evaluated using the ANSYS software through the FE-based approach rather than closed-form solutions. A mesh convergence assessment confirmed the precise simulation of the proposed airport runway FE model with the least element count and exact results. The proposed FE model's resultant eigenfrequencies and mode shapes have been evaluated while taking into consideration fifteen distinct vibrational modes, which confirms the accomplishment of the requirement of at least 90% effective mass of the structure's involvement. It was noticed that the eigenfrequency value raised with higher modes that showed complex deformation patterns such as a combination of flexure, translation, shear, and torsion. In addition, a case study was performed to categorize critical factors that affect the vibrational response of concrete pavement runways, which are the major findings of the present study. The present computational methodology was validated using a non-closed form numerical problem from the earlier study, and a good correlation between the present modeling approach and previous numerical and analytical results was found, which affirms the reliability of FE modeling in simulating complex structural behaviors, aiding in designing safer and more efficient runways. This study will also provide the base for further assessing the concrete pavement runway dynamic-transient response by incorporating the moving aircraft load.

During situations involving dangerous activities, such as armed robbery in public areas, surveillance systems often exhibit delays or inefficiencies in their prompt responding. To obviate the necessity for human involvement, there is a requirement for technology capable of autonomously detecting harmful objects, like firearms in surveillance footages. This article presents WeaponVision AI, an advanced software system that has the ability to accurately identify weapons in live feeds, recorded videos, and images. Moreover, this software has the capability to detect guns even under weak lighting circumstances. The deep learning architecture based on modified YOLOv7 was trained on a vast dataset assortment of 79,558 images of weapons, in developing the WeaponVision AI. The model exhibited satisfactory results following the training phase, achieving significant outcome metrics: a precision rate of 91.75% and a mean average precision of 92.15%. The efficacy of WeaponVision AI is showcased through its ability to accurately identify weapons across diverse environmental and visual conditions.

In recent years, there has been a significant increase in research studies that include the fabrication and characterization of metal matrix composites (MMCs) with unique features. This comprehensive review delves into the evolution and current status of copper MMCs (Cu‐MMCs) across various industrial sectors. Cu‐MMCs have garnered attention due to their remarkable properties, which include excellent thermal and electrical conductivity, corrosion resistance, and wear resistance. This study explores the fabrication processes, and intricate connections between microstructure and properties of Cu‐MMCs, which encompass ceramic and solid lubricants (SLs) reinforcements. The various types of reinforcement and fabrication methods are examined and highlighted advancements in designing compositions and optimizing microstructures during fabrication. Additionally, this study evaluates the friction and wear characteristics of self‐lubricating hybrid composites, providing insights into effective lubrication ranges and overall tribological behavior patterns. This review highlights that Cu‐MMCs demonstrate superior mechanical strength, wear resistance, and self‐lubricating properties due to ceramics and SLs reinforcements. The mechanisms underlying this behavior involve the formation of a protective transfer layer during sliding and effective lubrication provided by SLs, which reduces direct contact and facilitates smoother interactions between the mating surfaces. The review culminates in an outlook on the prospects of Cu‐MMCs, emphasizing the advantages conferred by their utilization.

In the present scenario, the development of efficient lithium-ion energy storage system-based electric vehicles has been turned into the focus as an effective alternative to conventional transportation systems. However, correspondingly associated potential risks of thermal failure confirm the need for an effective thermal management system to mitigate the adverse effects of excessive heat generated during operation. This work takes an opportunity to present a comparative performance assessment of response surface method (RSM) and artificial neural networks (ANN)-based predictive models to analyze the thermal behavior of oscillating heat pipes (OHP) filled with binary working fluids (acetone-DI water, acetone-methanol, and acetone-ethanol) with mixing ratio (1:1 ≤ MR ≤ 2:1) and volumetric filling ratio (30%≤FR ≤ 70%) as independent variables at different battery discharge rates (1C ≤ DR ≤ 2C). The thermal performance is predicted in terms of maximum cell temperature (MCT), average cell temperature (ACT), and thermal uniformity (TD). The acetone-DI water binary mixture was identified as the optimal working fluid among the fluids evaluated. The ACT, MCT, and TD were determined to be 38.3°C, 40°C, and 3.4°C, respectively, under a filling ratio (FR) of 70%, mixing ratio (MR) of 2:1, and a discharge rate (DR) of 2C. The prediction model performance is measured using correlation coefficient ( R ² ) value and average absolute prediction error (APE) to acknowledge the best fit. The results confirm the applicability of both the prediction models with the ANN model ( R ² -values ≥ 0.99998; average APE ≥ 0.0162%) slightly more accurate relative to the corresponding RSM approach ( R ² -values ≥ 0.9814; average APE ≥ 0.2342%).

A data-driven surrogate mimics the behavior of a black-box simulation model using selected input-output data points. Highly nonlinear models challenge many surrogate techniques in engineering design. This study proposes a reliable surrogate for training small-scale problems (two or three input variables). A combination of local and semi-global interpolators is introduced to approximate responses for new sample points in the design space. The local predictor uses a linear combination of two adjacent points, while the semi-global predictor employs K-means clustering, averaging samples in each cluster, and weighting clusters based on Euclidean distances. The trade-off between predictors is optimized using jackknife resampling error minimized via Genetic Algorithm (GA). Performance comparisons with Kriging, Radial Basis Function (RBF), and Support Vector Machine (SVM) confirm the proposed surrogate’s accuracy and robustness using five benchmark functions and two engineering problems. Results demonstrate superior interpolation of nonlinear functions with reduced error and improved robustness.

Rising atmospheric carbon dioxide (CO2) levels prompt concern for climate change, urging the immediate development of sustainable technologies to convert CO2 into valuable products. This study focused on developing a biocathode for bioelectrochemical CO2 reduction, utilizing Escherichia coli. The performance of the biocatalyst was assessed by varying wet cell mass from 0.0125 to 0.075 g/cm² under different operational conditions, including electrode potential (− 250 to − 1250 mV vs. Ag/AgCl) and methylene blue use as a mediator for electron transfer between electrode and microbial cells. Through optimization, the system achieved enhanced CO2 conversion rates and improved product selectivity. Beyond 50 h extended electrolysis period, consistent formic acid formation was observed with the highest Faradaic efficiency at 40 h reaction time. Produced formic acid shows a maximum concentration of 23.35 mM and a maximum Faradaic efficiency of 22.78% at − 750 mV vs. Ag/AgCl, using 0.05 g/cm² of wet cells. These findings hold significant promise for advancing technologies dedicated to CO2 sequestration and generating value-added products from CO2. Graphical abstract

The objective of this paper is to provide a comprehensive analysis of the index Whittaker transforms over Lebesgue spaces. It establishes Parseval relations and examines continuity properties, shedding light on their analytical characteristics and comparative merits.