Network Techniques Research Articles

Optimizing Ni (II) adsorption on Carica papaya fiber through experimental and advanced neural network prediction

ABSTRACT Detoxification of Ni (II) ion from the synthetic wastewater by Carica papaya fiber (CPF) has been attempted in batch method. FTIR analysis detected prominent functional groups that helped in binding Ni+2. The most efficient sorption was observed at pH 6 and with the increase in the ratio of biomass to Ni (II) ion strength, the removal efficiency improved. 2.5 g/L CPF powder removed 90.83% Ni (II) from 10 mg/L solution at a contact time of 120 min and at 30°C. Both linear and non-linear forms of Langmuir, Freundlich, and Temkin isotherms were attempted. Langmuir isotherm in both types matched well (R2 = 0.999 and 0.994) and the maximum adsorption capacity was 21.84 mg/g. The spontaneity of the reaction was confirmed by ΔG° values (−12.06 to −13.71 kJ/mol) and was more favorable at higher temperatures. The reaction was endothermic (ΔH° 83.03 kJ/mol) and random (ΔS° 83.07 J/mol/K). The desorption study proposed active reprocessing of CPF up to three cycles. The results were scaled up for use by industries. This study reflects sustainable eco-innovation and a cost-effective approach for locally based small-scale industries for environmental preservation. The feed-forward back propagation neural network technique was applied for efficient prediction of the Ni(II) removal efficiency.

Impact of Polystyrene Microplastics on Soil Properties, Microbial Diversity and Solanum lycopersicum L. Growth in Meadow Soils

The pervasive presence of microplastics (MPs) in agroecosystems poses a significant threat to soil health and plant growth. This study investigates the effects of varying concentrations and sizes of polystyrene microplastics (PS-MPs) on the Solanum lycopersicum L.’s height, dry weight, antioxidant enzyme activities, soil physicochemical properties, and rhizosphere microbial communities. The results showed that the PS0510 treatment significantly increased plant height (93.70 cm, +40.83%) and dry weight (2.98 g, +100%). Additionally, antioxidant enzyme activities improved across treatments for S. lycopersicum L. roots. Physicochemical analyses revealed enhanced soil organic matter and nutrient levels, including ammonium nitrogen, phosphorus, and effective potassium. Using 16S rRNA sequencing and molecular ecological network techniques, we found that PS-MPs altered the structure and function of the rhizosphere microbial community associated with S. lycopersicum L. The PS1005 treatment notably increased microbial diversity and displayed the most complex ecological network, while PS1010 led to reduced network complexity and more negative interactions. Linear discriminant analysis effect size (LEfSe) analysis identified biomarkers at various taxonomic levels, reflecting the impact of PS-MPs on microbial community structure. Mantel tests indicated positive correlations between microbial diversity and soil antioxidant enzyme activity, as well as relationships between soil physicochemical properties and enzyme activity. Predictions of gene function revealed that PS-MP treatments modified carbon and nitrogen cycling pathways, with PS1005 enhancing methanogenesis genes (mcrABG) and PS1010 negatively affecting denitrification genes (nirK, nirS). This study provides evidence of the complex effects of PS-MPs on soil health and agroecosystem functioning, highlighting their potential to alter soil properties and microbial communities, thereby affecting plant growth.

Recursive Quantum Relaxation for Combinatorial Optimization Problems

Quantum optimization methods use a continuous degree-of-freedom of quantum states to heuristically solve combinatorial problems, such as the MAX-CUT problem, which can be attributed to various NP-hard combinatorial problems. This paper shows that some existing quantum optimization methods can be unified into a solver to find the binary solution which is most likely measured from the optimal quantum state. Combining this finding with the concept of quantum random access codes (QRACs) for encoding bits into quantum states on fewer qubits, we propose an efficient recursive quantum relaxation method called recursive quantum random access optimization (RQRAO) for MAX-CUT. Experiments on standard benchmark graphs with several hundred nodes in the MAX-CUT problem, conducted in a fully classical manner using a tensor network technique, show that RQRAO not only outperforms the Goemans-Williamson and recursive QAOA methods, but also is comparable to state-of-the-art classical solvers. The code is available at https://github.com/ToyotaCRDL/rqrao.

Improved medical science image filtering with a novel neural network technique

Improved medical science image filtering with a novel neural network technique

Indonesia rupiah currency detection for visually impaired people using transfer learning VGG-19

People with visual impairments often face difficulties in determining the authenticity of paper money, which is a crucial skill to avoid fraud. The limitations of traditional methods, like blind codes for visually impaired people, require a more advanced and efficient solution. Previous methods of currency detection using Convolutional Neural Network (CNN) techniques, including the VGG-19 architecture, have often encountered challenges, particularly the long training times required. Therefore, we propose using transfer learning techniques and modifying the top layers of the VGG-19 model, known as fully connected layers, within a mobile application with audio feedback built using Android Studio. These modifications involve substituting the three fully connected layers with dense and flattened layers. We also implemented hyperparameter tuning, including adjusting the batch sizes and setting the number of epochs. The datasets used Indonesian Rupiah paper currency from the 2022 emission year, specifically Rp 50,000 and Rp 100,000 denominations. The best transfer learning VGG-19 model achieved a batch size of 32 and an epoch of 50, resulting in a high accuracy of 88%. Response speed testing with performance profiling on Android Studio showed an overall average response time of 458 ms. The main advantage of using transfer learning with the VGG-19 model is that it significantly reduces training time while still achieving high accuracy, differentiating this work from previous studies that relied on training from scratch, which is more time-consuming and resource-intensive. Therefore, this mobile app can be categorized as having a fast response time.

Application of convolutional neural networks for mobile internet forecasting in Colombia

The growth of mobile Internet has facilitated access to information by minimizing geographical barriers. For this reason, this paper forecasts the number of users, incomes, and traffic for operators with the most significant penetration in the mobile internet market in Colombia to analyze their market growth. For the forecast, the convolutional neural network (CNN) technique is used, combined with the recurrent neural network (RNN), long short-term memory network (LSTM), and gated recurrent unit (GRU) techniques. The CNN training data corresponds to the last twelve years. The results currently show a high concentration in the market since a company has a large part of the market; however, the forecasts show a decrease in its users and revenues and the growth of part of the competition. It is also concluded that the technique with the most precision in the forecasts is CNN-GRU.

Energy harvesting techniques for wireless sensor networks: A systematic literature review

Energy harvesting techniques for wireless sensor networks: A systematic literature review

Previsão de manutenção de pavimentos rígidos usando simulações numéricas e redes neurais artificiais

Abstract This work is concerned with the simulation of pavement behavior using the Finite Element Method (FEM) with the goal of predicting the appearance of cracks during its useful life. The developed method can help maintenance planning towards providing users with safety and comfort. The effects of both traffic loads and temperature variations are considered. Artificial Intelligence (AI) tools are adopted to reduce the time necessary for the interpretation of simulation results for the development of an efficient pavement management system. The developed rigid pavement management system uses the Artificial Neural Networks (ANN) technique for the prediction of both pavement response to fatigue accumulation and the behavior of the modeled pavement with a high degree of precision. The networks showed 4.5% and 3.6% square errors for positive and negative temperature gradients and 95% and 97% data correlation, respectively. The SR stress ratio obtained by the developed ABAQUS model is 0.614 and the value provided by the pavement management system is 0.648. The developed methodology was applied to a section of highway BR-101/NE, between the states of Paraiba PR and Pernambuco PE, in Brazil. Comparing the results with the limits provided by National Cooperative Highway Research Program (NCHRP) for traffic in the study, this pavement would require maintenance every six months. The maximum values of tensile stress due to bending were obtained by combining the positive temperature gradient with the traffic axis load at the edge during the summer months (December to March) due to the higher thermal gradient values.

Enhancing Malicious User Recognition Using Coot Optimization Algorithm with Bayesian Belief Network for Cognitive Radio Networks

As a dynamic paradigm, Cognitive radio networks (CRNs) in wireless transmission enable devices to intelligently adapt their communication parameter based on real-world spectrum availability. Spectrum sensing lies at the core of CRNs, where nodes continue to monitor the spectrum for underutilized or unused band detection. However, the presence of malicious users (MUs) has a significant impact reliability and performance of the network. MUs detection is indispensable to prevent interference or unauthorized access and ensure network integrity. Advanced techniques combining game theory, machine learning, and signal processing are used for effectively identifying and mitigating malicious activities. CRNs can ensure efficient spectrum utilization and enhance security in heterogeneous and dynamic environments by incorporating robust MU detection systems into spectrum sensing protocols. This article presents a Malicious User Recognition using the Coot Optimization Algorithm with Bayesian Belief Network (MUR-COABBN) technique for CRN. The MUR-COABBN technique exploits metaheuristics with a Bayesian machine-learning method for the classification of the MUs in the CRN. In the MUR-COABBN technique, the COA is initially used to choose better feature subsets. Moreover, the detection of MUs can be performed by the use of BBN. Finally, the parameter tuning of the BBN model is carried out using an improved seeker optimization algorithm (ISOA). The experimental evaluation of the MUR-COABBN technique takes place with respect to distinct aspects. The experimentation outcomes implied the improved performance of the MUR-COABBN methodology with other methods under distinct measures. Therefore, the MUR-COABBN model can effectually and accurately improve security in the CRN.

Water pollution and water quality assessment and application of criterion impact loss (CILOS), geographical information system (GIS), artificial neural network (ANN) and decision-learning technique in river water quality management: An experiment on the Mahanadi catchment, Odisha, India

Water pollution and water quality assessment and application of criterion impact loss (CILOS), geographical information system (GIS), artificial neural network (ANN) and decision-learning technique in river water quality management: An experiment on the Mahanadi catchment, Odisha, India

Exploration of Arrhenius activation energy and thermal radiation on MHD double-diffusive convection of ternary hybrid nanofluid flow over a vertical annulus with discrete heating

The primary objective of this article is to examine the effect of discrete heating on MHD double-diffusive convection and thermal radiation of ternary hybrid nanofluid flow heat and mass transfer in a vertical cylindrical annulus along with Arrhenius activation energy and chemical reaction. In this study, the cavity inner wall has two distinct flush-mounted heat sources, while the outer wall is isothermally cooled at a lower temperature. The top and bottom walls are thermally insulated. The ensuing equations that govern the physical framework are solved using the implicit Crank-Nicholson finite difference technique. As the heater advances upward, the flow intensity decreases, leaving a part of the fluid static at the bottom of the cylinder. Because more heat induces high buoyant flow in the annulus, the absolute value of axial velocity and wall temperature rises as the length of the heat source rises. Enhancing the activation energy parameter values drops the Arrhenius energy function, elevating the pace of the generative chemical process and hence the concentration. Increasing the thermal radiation parameter lowers the surface heat flux while enhancing the nanofluid temperature. The Brownian motion parameter corresponds to the random motion of nanoparticles in a fluid, and this irregular movement augments the collision of nanoparticles with fluid particles, causing the particle's kinetic energy which leads to thermal energy and hence increases temperature, a 10% increment in Brownian motion parameter leads to 26% rise in the temperature. Also, the heat and mass transfer characteristics are forecasted and analyzed by considering the Levenberg–Marquardt backpropagating artificial neural network technique.

Artificial Intelligence-Based Direct Power Control for Power Quality Improvement in a WT-DFIG System via Neural Networks: Prediction and Classification Techniques

This paper discusses the improvement of power quality injected into the AC grid. This approach is achieved by enhancing the quality of injected power signals and mastering the active and reactive power exchanged between the DFIG based wind turbine (WT-DFIG) and the electrical grid, resulting in an improvement of the overall system performance and efficiency. This study includes all WT-DFIG operating modes, successively and continuously, as well as all local reactive power compensation modes. Therefore, novel control strategies are proposed in this paper for wind energy conversion systems based on artificial intelligence techniques. These techniques include Neural Network Prediction (PNN-DPC) and Classification (CNN-DPC). They aim to eliminate the drawbacks and difficulties associated with conventional Direct Power Control (C-DPC), while retaining its advantages. The paper also provides a thorough explanation of the mathematical models for neural network techniques and WT-DFIG system models. The MATLAB/Simulink environment is used to investigate the performance of the proposed techniques under different conditions and operating modes related to different scenarios. The results reveal a significant reduction in the ripple of the generated active power and the compensated local reactive power, better quality of the generated signal currents and a remarkable reduction in the current total harmonic distortion (THD). Furthermore, compared to C-DPC, PNN-DPC achieves a reduction of 72.07% in active power ripples, 77.07% in reactive power ripples, and 76.79% in current Total Harmonic Distortion (THD). CNN-DPC shows similar improvements with 72.04%, 77.13%, and 76.54% of reductions respectively. In addition, CNN-DPC slightly outperforms PNN-DPC. Nevertheless, both proposed control techniques show significant improvements in all characteristics compared to other methods. Consequently, the proposed control strategies indicate that artificial intelligence has the potential to improve the power quality and performance of wind power conversion system.

Bayesian Network approach in analyzing the sustainability of the cultural industry ‘the sacred’ Gringsing Weaving

Bali is a popular tourist destination in the world, and the main entry point for foreign tourists to Indonesia. Bali is also an area in Indonesia that is famous for producing woven fabrics, which have various characteristics in each region. Gringsing weaving is one of the traditional Balinese fabrics from Tenganan Village which is considered sacred and its manufacture takes quite a long time, and is carried out with special techniques that are very difficult. This study aims to analyze and map factors related to the sustainability of the Gringsing weaving cultural industry, using the Bayesian Network approach. The results of the FGD with related stakeholders mapped the structure that forms Gringsing industrial sustainability, such as income, social capital, incentives, natural capital, custom law, and traditional institutions. Further analysis was carried out using the Bayesian Network technique and GeNIe tools. In forming the structure of thinking about the sustainability of the Gringsing Weaving industry, the related factors include culture, traditional institutions, natural capital, social capital, to economic factors, such as income and special incentives from the government. Specifically, the role of traditional institutions and government (through special incentives) was analyzed, and it was found that both factors can increase the probability of the sustainability of the gringsing weaving industry. The results of the sensitivity analysis also show that the sustainability of the Gringsing weaving industry is highly influenced (sensitive) to the increasing role of traditional institutions.

Integration GSTARIMA with deep neural network to enhance prediction accuracy on rainfall data

This study aimed to improve rainfall prediction accuracy by integrating spatio-temporal Generalized Autoregressive Integrated Moving Average (GSTARIMA) with Deep Neural Network (DNN) techniques. Accurate rainfall forecasting is vital for regions like West Java, which face risks of flooding and landslides during heavy rains in December, January and February. The research focused on enhancing monthly rainfall prediction using GSTARIMA model residuals, transformed from daily data, with a spatial resolution of 0.5° × 0.625°. The GSTARIMA-DNN model was developed to reduce prediction errors and closely align with actual rainfall values. The DNN architecture utilized a Multilayer Perceptron (MLP) with eight input nodes for spatio-temporal residuals, two hidden layers with Leaky ReLU (LReLU) activation functions and a linear output layer for regression. The integration of GSTARIMA and DNN significantly improved prediction accuracy compared to traditional methods. Specifically, the GSTARIMA(0,1,1)-DNN model achieved a 7.28% Mean Absolute Percentage Error (MAPE) at four key locations (Tasikmalaya, Cirebon, Majalengka and Garut) among ten locations with the same lag order. This hybrid model effectively managed complex temporal and spatial data, underscoring the benefits of combining spatio-temporal models with DNN for enhanced rainfall forecasting. This approach offers valuable insights for advancing precise spatio-temporal models in various research fields.

A quantitative benchmark of neural network feature selection methods for detecting nonlinear signals

Classification and regression problems can be challenging when the relevant input features are diluted in noisy datasets, in particular when the sample size is limited. Traditional Feature Selection (FS) methods address this issue by relying on some assumptions such as the linear or additive relationship between features. Recently, a proliferation of Deep Learning (DL) models has emerged to tackle both FS and prediction at the same time, allowing non-linear modeling of the selected features. In this study, we systematically assess the performance of DL-based feature selection methods on synthetic datasets of varying complexity, and benchmark their efficacy in uncovering non-linear relationships between features. We also use the same settings to benchmark the reliability of gradient-based feature attribution techniques for Neural Networks (NNs), such as Saliency Maps (SM). A quantitative evaluation of the reliability of these approaches is currently missing. Our analysis indicates that even simple synthetic datasets can significantly challenge most of the DL-based FS and SM methods, while Random Forests, TreeShap, mRMR and LassoNet are the best performing FS methods. Our conclusion is that when quantifying the relevance of a few non linearly-entangled predictive features diluted in a large number of irrelevant noisy variables, DL-based FS and SM interpretation methods are still far from being reliable.

Leveraging fuzzy embedded wavelet neural network with multi-criteria decision-making approach for coronary artery disease prediction using biomedical data

Coronary artery disease (CAD) is the main cause of death. It is a complex heart disease that is linked with many risk factors and a variety of symptoms. In the past few years, CAD has experienced a remarkable growth. Prompt risk prediction of CAD would be capable of decreasing the death rate by permitting timely and targeted treatments. Angiography is the most precise CAD diagnosis technique; however, it has several side effects and is expensive. Multi-criteria decision-making approaches can well perceive CAD by analysing main clinical indicators like ChestPain type, ST_Slope, and HeartDisease presence. By assessing and evaluating these factors, the model improves diagnostic accuracy and aids informed clinical decisions for quick CAD detection. Mainly machine learning (ML) and deep learning (DL) use plentiful models and algorithms, which are commonly employed and very useful in exactly detecting the CAD within a short time. Current studies have employed numerous features in gathering data from patients while using dissimilar ML and DL models to attain results with high accuracy and lesser side effects and costs. This study presents a Leveraging Fuzzy Wavelet Neural Network with Decision Making Approach for Coronary Artery Disease Prediction (LFWNNDMA-CADP) technique. The presented LFWNNDMA-CADP technique focuses on the multi-criteria decision-making model for predicting CAD using biomedical data. In the LFWNNDMA-CADP method, the data pre-processing stage utilizes Z-score normalization to convert an input data into a uniform format. Furthermore, the improved ant colony optimization (IACO) method is used for electing an optimum sub-set of features. Furthermore, the classification of CAD is accomplished by utilizing the fuzzy wavelet neural network (FWNN) technique. Finally, the hyperparameter tuning of the FWNN model is accomplished by employing the hybrid crayfish optimization algorithm with the self-adaptive differential evolution (COASaDE) technique. The simulation outcomes of the LFWNNDMA-CADP approach are investigated under a benchmark database. The experimental validation of the LFWNNDMA-CADP approach portrayed a superior accuracy value of 99.49% over existing techniques.

Urban Water-Energy consumption Prediction Influenced by Climate Change utilizing an innovative deep learning method

The growing global demand for water and energy has created an urgent necessity for precise forecasting and management of these resources, especially in urban regions where population growth and economic development are intensifying consumption. Shenzhen, a rapidly expanding megacity in China, exemplifies this trend, with its water and energy requirements anticipated to rise further in the upcoming years. This research proposes an innovative Convolutional Neural Network (CNN) technique for forecasting water and energy consumption in Shenzhen, considering the intricate interactions among climate, socio-economic, and demographic elements. The proposed approach integrates a CNN model with an Enhanced Gorilla Troops Optimization (EGTO) algorithm to demonstrate superior performance compared to other leading methods in terms of accuracy and reliability. The results show a strong correlation between the simulated and observed data, with a correlation coefficient of 0.87 for water consumption and 0.91 for energy consumption, indicating a high level of agreement between the simulated and real-world data. Also, it is indicated that the new technique can accurately forecast water and energy consumption, achieving a mean absolute error (MAE) of 0.63 and a root mean square error (RMSE) of 0.58, respectively. The research indicates that the suggested approach can promote policymakers and stakeholders in making well-informed decisions by delivering precise predictions of water and energy usage. This, in turn, can facilitate better resource distribution, minimize waste, and promote greater sustainability. The study emphasizes the necessity of incorporating climate change and socio-economic factors into the forecasting process and showcases the proposed method’s potential to aid decision-making in this domain.

Application of human-computer interaction technology integrating biomimetic vision system in animation design with a biomechanical perspective

The combination of Human-Computer Interaction (HCI) technology with biomimetic vision systems has transformational potential in animation design, particularly by incorporating biomechanical principles to create immersive and interactive experiences. Traditional animation approaches frequently lack sensitivity to real-time human motions, which can restrict engagement and realism. This study addresses this constraint by creating a framework that uses Virtual Reality (VR) and Augmented Reality (AR) to generate dynamic settings that include a variety of human activities, informed by biomechanical analysis. A biomimetic vision system is used to record these motions with wearable sensors, allowing for precise monitoring of user activity while considering biomechanical factors such as joint angles, force distribution, and movement patterns. The recorded data is preprocessed using Z-score normalization methods and extracted using Principal Component Analysis (PCA). This study proposed an Egyptian Vulture optimized Adjustable Long Short-Term Memory Network (EVO-ALSTM) technique for motion classification, specifically tailored to recognize biomechanical characteristics of human movements. Results demonstrate a significant improvement in precision (93%), F1-score (91%), accuracy (95%), and recall (90%) for the motion recognition system, highlighting the effectiveness of biomechanical insights in enhancing animation design. The findings indicate that integrating real-time biomechanical data into the animation process leads to more engaging and realistic user experiences. This study not only advances the subject of HCI but also provides the framework for future investigations into sophisticated animation technologies that use biomimetic and biomechanical systems.

Digital Image Processing (DIP) and Generative Adversarial Networks (GANs) Techniques for Improvement Low-Resolution Face Recognition

Digital Image Processing (DIP) and Generative Adversarial Networks (GANs) Techniques for Improvement Low-Resolution Face Recognition

A Survey of Emerging Techniques for Large Networks of Virtual Local Area Networks (VLANs) with Benefits and Limitations

The numerous benefits of wireless communication have led to its meteoric rise in popularity as a network type, yet there have been numerous obstacles to enhancing its performance. An alternative method that allows a network administrator to construct a logical network out of a physical network is VLAN, which stands for Virtual Local Area Network. Thus, VLANs are fundamental technology for enhancing network performance, capacity and security. VLANs are a means of breaking large networks into smaller broadcast domains, managing broadcast traffic, controlling congestion, and providing increased levels of security. This paper explores the emerging techniques in managing and optimising VLANs within large-scale network environments. VLANs are essential for improving network performance, security, and scalability by the logical segmentation of networks into smaller, isolated broadcast domains. This survey also outlines trends in future VLAN technologies such as Software Defined Networking (SDN), VXLAN, and Big Data and Analytics. Also, the paper covers the uses, advantages, and disadvantages of VLANs, including enhanced security, scalability, and economy, though admitting to issues of VLAN configuration difficulty and security threats. In the paper, the authors discussed the future trends including the integration of multi-cloud, network slicing coupled with 5G, and integration of blockchain for VLAN security.