Network Applications Research Articles

Vacancy-driven resistive switching behavior based on wafer-scale MoSe2 artificial synapses

Recently, researches have revealed that synaptic transistors based on two-dimensional materials (2DMs) have substantial advantages and potential for modeling neural synapses. However, the mechanism of artificial synaptic device with 2DMs as conducting channels remains to be further clarified. Here, high-quality wafer-scale MoSe2 nanosheets were achieved by ALD-CVD process, where further 23 × 23 × 4 synaptic arrays were fabricated in situ. Notably, the outstanding MoSe2 synaptic device exhibits an on/off ratio of 3.6 × 107 with hysteresis ratio of 6.8k and simulates crucial biological synaptic behaviors such as paired pulse facilitation and excitatory postsynaptic currents. The mechanism of Se vacancy induction and horizontal diffusion driven is revealed based on high-resolution transmission electron microscopy characterization with first-principles calculations, which suggests a possible mechanistic explanation for the resistive switching behavior of 2DMs transistors. Artificial synaptic devices with 2DMs based on the working mechanism of vacancy diffusion are capable of satisfying the needs of complex neural network applications.

Malicious Nodes Detection and Avoidance Using Trust-based Routing in Critical Data Handling Wireless Sensor Network Applications

Malicious Nodes Detection and Avoidance Using Trust-based Routing in Critical Data Handling Wireless Sensor Network Applications

The use of CNNs in VR/AR/MR/XR: a systematic literature review

This study offers a systematic literature review on the application of Convolutional Neural Networks in Virtual Reality, Augmented Reality, Mixed Reality, and Extended Reality technologies. We categorise these applications into three primary classifications: interaction, where the networks amplify user engagements with virtual and augmented settings; creation, showcasing the networks’ ability to assist in producing high-quality visual representations; and execution, emphasising the optimisation and adaptability of apps across diverse devices and situations. This research serves as a comprehensive guide for academics, researchers, and professionals in immersive technologies, offering profound insights into the cross-disciplinary realm of network applications in these realities. Additionally, we underscore the notable contributions concerning these realities and their intersection with neural networks.

Simulation of synaptic properties of ferroelectric memory capacitors and neural network applications

In this work, the electrical properties and synaptic characteristics of hafnium oxide-based ferroelectric memory capacitor with metal - ferroelectric layer - metal (MFM) structure were simulated using TCAD (technology computer aided design) software. Based on the synaptic potentiation/depression characteristics of the simulated memory capacitor, a multilayer perceptron (MLP) network was constructed, and the recognition accuracy and convergence speed of the MLP network in the MNIST recognition task were simulated, and the feasibility of the ferroelectric memory capacitor synaptic device for real neural network operation was analyzed. The results show that the recognition accuracy of the MLP network reaches 93% and stabilizes after 50 iterations of training, and the recognition accuracy of the MLP network is already at a high usable level after a smaller number of training times of 20, which suggests that the synaptic plasticity of the ferroelectric memory capacitor has a good potential for the practical application of the weight updating of the MLP network.

Research on weathering recognition model of ancient glass based on three-layer neural network

In this paper, a three-layer neural network-based weathering recognition model for ancient glass is proposed to cope with the limitations of traditional weathering analysis methods in dealing with complexity and large-scale data. By constructing a multi-layer feed-forward neural network model containing input, hidden and output layers, this paper achieves high-precision classification of the weathering state of ancient glass. The experimental results show that the accuracy of the model reaches 98.0% and 100% on the training set and test set, respectively, which proves the effectiveness and efficiency of the method in the recognition of ancient glass weathering. This study demonstrates the potential of neural network application in cultural relics conservation and provides a theoretical basis for further optimisation of cultural relics identification techniques.

Dimensionality reduction using neural networks for lattice-based cryptographic keys

Post Quantum Cryptography has received an increasing amount of active research. This has been made prominent by the ever-growing field of quantum computing which poses a formidable threat to the modern cybersecurity landscape. Recently, post quantum schemes have been standardized for adoption into existing security services and protocols. In comparison to contemporary cryptographic schemes, these approaches are lagging behind in terms of performance with regards to functional speed and package sizes. A study into the various applications of neural networks used in classical and post quantum cryptographic schemes has been explored to demonstrate their different possible applications within the various fields of cybersecurity. The main contribution of this work is to investigate the feasibility of dimensionality reduction using the autoencoder neural network on lattice-based keys generated by the Kyber key encapsulation mechanism scheme. Moreover, this work also presents a comparative analysis of the different implemented autoencoder models to showcase their relative performance.

Rhizobium etli CFN42 and Sinorhizobium meliloti 1021 bioinformatic transcriptional regulatory networks from culture and symbiosis.

Rhizobium etli CFN42 proteome-transcriptome mixed data of exponential growth and nitrogen-fixing bacteroids, as well as Sinorhizobium meliloti 1021 transcriptome data of growth and nitrogen-fixing bacteroids, were integrated into transcriptional regulatory networks (TRNs). The one-step construction network consisted of a matrix-clustering analysis of matrices of the gene profile and all matrices of the transcription factors (TFs) of their genome. The networks were constructed with the prediction of regulatory network application of the RhizoBindingSites database (http://rhizobindingsites.ccg.unam.mx/). The deduced free-living Rhizobium etli network contained 1,146 genes, including 380TFs and 12 sigma factors. In addition, the bacteroid R. etli CFN42 network contained 884 genes, where 364 were TFs, and 12 were sigma factors, whereas the deduced free-living Sinorhizobium meliloti 1021 network contained 643 genes, where 259 were TFs and seven were sigma factors, and the bacteroid Sinorhizobium meliloti 1021 network contained 357 genes, where 210 were TFs and six were sigma factors. The similarity of these deduced condition-dependent networks and the biological E. coli and B. subtilis independent condition networks segregates from the random Erdös-Rényi networks. Deduced networks showed a low average clustering coefficient. They were not scale-free, showing a gradually diminishing hierarchy of TFs in contrast to the hierarchy role of the sigma factor rpoD in the E. coli K12 network. For rhizobia networks, partitioning the genome in the chromosome, chromids, and plasmids, where essential genes are distributed, and the symbiotic ability that is mostly coded in plasmids, may alter the structure of these deduced condition-dependent networks. It provides potential TF gen-target relationship data for constructing regulons, which are the basic units of a TRN.

Application of Artificial Intelligence in the Management of Coagulation Treatment Engineering System

In this paper, the application of artificial intelligence, especially neural networks, in the field of water treatment is comprehensively reviewed, with emphasis on water quality prediction and chemical dosage optimization. It begins with an overview of machine learning and deep learning concepts relevant to water treatment. Key advances and challenges in using neural networks for coagulation processes are thoroughly analyzed, including the automation of coagulant dosing, dosage level optimization, and efficiency comparisons of modeling approaches. Applications of neural networks in predicting pollutant levels and supporting water quality monitoring are explored. The review identifies avenues for improving coagulation-based modeling with neural networks, such as enhancing data quality, employing feature engineering, refining model selection criteria, and improving cross-validation methods. The necessity of continuous monitoring and adaptive optimization strategies is emphasized. Challenges such as the complexity of coagulation processes, feedback control signal acquisition, and model adaptability from simulations to real-world settings are discussed. Cost control and resource management in water treatment are also highlighted, emphasizing the optimized chemical dosage to reduce expenses while maintaining water quality compliance. In summary, this review provides valuable insights into the current state of neural network applications in water treatment and highlights key areas for further research and development. Integrating AI into coagulation processes has the potential to enhance the efficiency and sustainability of drinking water treatment.

Performance investigation of a VLC-PDM based UWOC system under adverse underwater conditions with varying chlorophyll levels

Underwater Wireless Optical Communication (UWOC) has emerged as a promising solution to meet the growing demand for high-speed, reliable communication in marine environments. The limitations of traditional acoustic and radio frequency (RF) communication methods in underwater settings, such as low data rates, high latency, and susceptibility to environmental conditions, have propelled the exploration of optical communication technologies. This work introduces a Visible Light Communication (VLC)-based UWOC system employing Polarization Division Multiplexing (PDM) to enhance data transmission rates and system performance. The proposed system capitalizes on the advantages of VLC, including high bandwidth, low latency, and immunity to electromagnetic interference, to achieve efficient underwater communication. By incorporating PDM, the system effectively doubles the channel capacity without increasing the bandwidth, achieving a data transmission rate of 1Gbps per channel. The performance of the proposed system was evaluated under various underwater conditions, including freshwater, clear ocean water, coastal water, and turbid ocean water with different chlorophyll concentrations. The results demonstrated the system's ability to maintain a low Bit Error Rate (BER) across diverse marine environments, showcasing its potential for real-world applications in underwater communication networks.

High-Linearity Ta2O5 Memristor and Its Application in Gaussian Convolution Image Denoising.

In the image Gaussian filtering process, convolving with a Gaussian matrix is essential due to the numerous arithmetic computations involved, predominantly multiplications and additions. This can heavily tax the system's memory, particularly with frequent use. To address this issue, a W/Ta2O5/Ag memristor was employed to substantially mitigate the computational overhead associated with convolution operations. Additionally, an interlayer of ZnO was subsequently introduced into the memristor. The resulting Ta2O5/ZnO heterostructure layer exhibited improved linearity in the pulse response, which enhanced linearity facilitates easy adjustment of the conductance magnitude through a linear mapping of the number of pulses and the conductance. Subsequently, the conductance of the W/Ta2O5/ZnO/Ag bilayer memristor was employed as the weights for the convolution kernel in convolution operations. Gaussian noise removal in image processing was achieved by assembling a 5 × 5 memristor array as the kernel. When denoising was performed using memristor arrays, compared to denoising achieved through Gaussian matrix convolution, an average loss of less than 5% was observed. The provided memristors demonstrate significant potential in convolutional computations, particularly for subsequent applications in convolutional neural networks (CNNs).

Sorting Permutations on an n − Broom

With applications in computer networks, robotics, genetics, data center network optimization, cryptocurrency exchange, transportation and logistics, cloud computing, and social network analysis, the problem of sorting permutations on transposition trees under various operations is highly relevant. The goal of the problem is to sort or rearrange the markers in a predetermined order by swapping them out at the vertices of a tree in the fewest possible swaps. Only certain classes of transposition trees, like path, star, and broom, have computationally efficient algorithms for sorting permutations. In this paper, we examine the so-called n−broom transposition trees. A single broom or simply a broom is a spanning tree formed by joining the center of the star graph with one end of the path graph. A generalized version of a broom known as an n−broom is created by joining the ends of n brooms to one vertex, known as the n−broom center. By using the idea of clear path markers, we present a novel algorithm for sorting permutations on an n−broom for n>2 that reduces to a novel 2−broom algorithm and that further reduces to two instances of a 1−broom algorithm. Our single-broom algorithm is similar to that of Kawahara et al.; however, our proof of optimality for the same is simpler.

KTz Neuron in Electromagnetic Radiation with a Novel Flux-Controlled Memristor

In this paper, a novel discrete flux-controlled generalized memristor is proposed to be used in Electromagnetic Radiation (EMR) KTz neuron. The introduction of flux-controlled memristor can simulate the behavior changes of the neuron under EMR. The stability of flux-controlled memristor EMR KTz Neuron (EKN) map is discussed, and the fixed point of the system is characterized. With the change of parameters, the complex and rich dynamical characteristics of the EKN map are demonstrated. The firing patterns under different initial values are also investigated, and discharge waveforms similar in general shape and structure to human heart and brain disease phenomena are analyzed. Finally, the hardware implementation is verified on the Digital Signal Processing (DSP) platform. The results show that these findings are of great value for the practical applications of neural networks, with a theoretical basis provided.

Optimizing Sales and Inventory Management with Machine Learning: Applications of Neural Networks and Random Forests

Optimizing Sales and Inventory Management with Machine Learning: Applications of Neural Networks and Random Forests

Computer Network Design and Applications for West Java Legislative Elections

Legislative elections or Pileg are an annual schedule that is routinely held in Indonesia for five years. The concept of technology 4.0 has a huge opportunity to be applied in the legislative election process in this modern era. Legislative elections that use digital devices connected in such a way have more advantages than legislative elections that use conventional methods. The application of technology 4.0 in legislative elections has advantages in terms of effectiveness, data processing speed, and costs. The design of the legislative election system is made using a network where each polling station will be provided with one server to accommodate votes at the polling station. Then the server synchronizes the data to a central server to calculate the final results obtained. The server is built using the text-based Ubuntu Linux operating system. The application used is web-based, with the programming languages used being HTML, PHP, CSS, and Javascript. The test data shows that the time required for the system to complete each stage of the legislative election process is relatively faster than conventional legislative elections. So it can be concluded that the design of the legislative election system has met the requirements to accelerate the legislative election process. This research uses an SDLC system with a prototype model. Tests are carried out using BlackBox testing to test the suitability of the application with application access.

Low-energy and tunable LIF neuron using SiGe bandgap-engineered resistive switching transistor

We have proposed leaky integrate-and-fire (LIF) neuron having low-energy consumption and tunable functionality without external circuit components. Our LIF neuron has a simple configuration consisting of only three components: one bandgap-engineered resistive switching transistor (BE-RST), one capacitor, and one resistor. Here, the crucial point is that BE-RST with a silicon–germanium heterojunction possesses an amplified hysteric current switching with a low latch-up voltage due to improved hole storage capability and impact ionization coefficient. Therefore, the proposed neuron utilizing BE-RST requires an energy consumption of 0.36 pJ/spike, which is approximately six times lower than 2.08 pJ/spike of pure silicon-RST based neuron. In addition, the spiking properties can be tuned by modulating the leakage rate and threshold through gate bias, which contributes to energy-efficient sparse-activity and high learning accuracy. As a result, our proposed neuron can be a promising candidate for executing various spiking neural network applications.

Physics-constrained deep learning approach for solving inverse problems in composite laminated plates

The applications of physics-informed neural networks (PINNs) in material parameters identification of composite laminates are currently research highlights. We present an efficient physics-constrained deep learning approach based on PINNs for solving material parameters of composite laminates. We explain the details of incorporating first-order shear deformation theory as physical information into designed loss functions. The performance of proposed approach is demonstrated through a numerical example of idealized composite laminated plates under uniform pressure conditions. The computational errors of material parameters are less than 0.1%. Moreover, the applicability of transfer learning (TL) has been illustrated for inversion of material parameters with few datasets. The ambition is to improve the identification efficiency of carbon and glass fiber reinforced plies material parameters across various loading conditions. Intensive studies are conducted to compare convergence time between with and without TL through a numerical example under thermal-force coupling conditions. The comparison shows PINNs with TL obtain average error of 0.144% and 0.077% in carbon and glass fiber reinforced plies, which is better than without TL of 0.472% and 0.284%, respectively. The proposed work emphasizes the effective application of PINNs-based approach that combines first-order deformation and data-driven solution features for inverse solving material parameters of composite laminates.

Artificial Intelligence System Based on Wireless Network and Optical Sensor Recognition Application in Museum Interactive VR Design

Artificial Intelligence System Based on Wireless Network and Optical Sensor Recognition Application in Museum Interactive VR Design

Leveraging Time-Critical Computation and AI Techniques for Task Offloading in Internet of Vehicles Network Applications

Leveraging Time-Critical Computation and AI Techniques for Task Offloading in Internet of Vehicles Network Applications

SFDM: A time-synchronization-free detection mechanism for mobile underwater acoustic sensor networks

Detection and localization are important applications of underwater acoustic sensor networks, and the use of passive listening arrays composed of multiple underwater unmanned platforms to achieve passive detection and localization of targets has been a hot research topic in recent years. However, most schemes based on mobile listening arrays only aim to improve localization accuracy, with less consideration of the high communication and time overhead of formation maintenance and the high cost of time synchronization among nodes. To reduce the cost while ensuring localization accuracy, this paper proposes a time-synchronization-free detection mechanism for mobile underwater acoustic sensor arrays. Initially, the nodes cruise underwater until they passively detect a target signal. Subsequently, a pseudo-synchronization and rough self-calibration process is triggered among the nodes, which relies on the communication among the nodes. The communication stage is based on a time-synchronization-free detection protocol. An improved sound speed stratification effect compensation algorithm and an unscented particle filter algorithm are proposed to improve the calibration accuracy of the array. Finally, the Time Difference of Arrival method is utilized to locate the target position. Simulation and sea trial results confirm the feasibility and accuracy of the proposed mechanism. In addition, the mechanism is effective in reducing hardware costs.

A low‐profile, wideband, circularly polarized, slotted, U‐shaped textile antenna for W‐BAN applications

AbstractWe propose a compact, wideband, U‐shaped, circularly polarized textile antenna (UCPTA) resonating at 2.45 GHz for wireless body area network (W‐BAN) applications. The radiator of the antenna is printed on the top of the substrate (size of mm3; ) and has a full ground plane on the other side. The proposed antenna's substrate consists of a single layer of denim (), and conductive copper tape was used to fabricate the patch. The antenna has a low‐profile design that is suitable for wearable applications. A small slot was introduced at the center of the U‐shaped patch to enhance the 3 dB axial ratio bandwidth (ARBW) up to 55.17%. The proposed antenna radiates with a 10 dB return loss (RL), impedance bandwidth (IBW) of 64.40%, and a simulated gain of 5.1 dBi. The wearable textile antenna with broadside radiation response is appropriate for off‐body communications. The prototype of UCPTA is fabricated for experimental validation. The measured results for the fabricated antenna were in good agreement with the simulated results. Further, it has been demonstrated that the antenna's performance is stable during structural deformation and human body loading. This textile antenna covers an ARBW in the frequency band of 2.1 GHz–3.72 GHz and has a broad IBW (2 GHz–3.9 GHz) for LTE (2.17 GHz), MBAN (2.360 GHz–2.4 GHz), WPAN (2.395 GHz–2.4 GHz), and WLAN (2.4 GHz–2.482 GHz) applications.