Adaptive Network Research Articles

Low-complexity end-to-end deep learning framework for 100G-PON

End-to-end learning allows communication systems to achieve optimal performance compared with conventional blockwise structure design. By modeling the channel with neural networks and training the transmitter and receiver on this differentiable channel, the whole system can be jointly optimized. However, in existing schemes, channel modeling methods, such as the generative adversarial network and long short-term memory network, have complex architectures and cannot track channel changes, leading to less effective end-to-end learning. Meanwhile, the complexity of neural networks deployed at the transmitter and receiver is too high for practical applications. In this work, we propose an efficient and low-complexity end-to-end deep learning framework and experimentally validate it on a 100G passive optical network. It uses a noise adaptation network to model channel response and noise distribution and employs offline pretraining and online tracking training to improve the efficiency and accuracy of channel modeling. For the transmitter, it consists of a pattern-dependent look-up table (PDLUT) based on a neural network (NN-PDLUT) with a single convolutional layer. Further, the receiver is also an NN with a single convolutional layer; thus, the end-to-end signal processing is extremely simple. The experimental results show that end-to-end learning improves the receiver sensitivity by 0.85 and 1.59 dB compared with receiver-only equalization based on Volterra nonlinear equalization (VNLE) and joint equalization based on a PDLUT and a feed-forward equalizer, respectively. Moreover, the number of multiply–accumulate operations consumed by the transmitter and receiver in the end-to-end learning scheme is reduced by 75.7% compared with VNLE-based receiver-only equalization.

MFHARFNet: Multi-branch feature hybrid and adaptive receptive field network for image segmentation

Abstract Accurate segmentation of medical images is crucial for disease diagnosis and understanding disease changes. Deep learning methods, utilizing encoder-decoder structures, have demonstrated cutting-edge performance in various medical image segmentation tasks. However, the pooling operation in the encoding stage results in feature loss, which makes the network lack the ability to fuse multi-scale information at different levels, hinders its effective perception of multi-scale information, and leads to poor segmentation performance. Drawing inspiration from the U-shaped network, this study introduces a multi-branch feature hybrid attention and adaptive receptive field network (MFHARFNet) for medical image segmentation. Building upon the encoder-decoder framework, we initially devise a multi-branch feature hybrid attention module (MFHAM) to seamlessly integrate feature maps of varying scales, capturing both fine-grained features and coarse-grained semantics across the entire scale. Furthermore, we redesign the skip connection to amalgamate feature information from different branches in the encoder stage and efficiently transmit it to the decoder, providing the decoder with global context feature maps at different levels. Finally, the adaptive receptive field (ARF) module is introduced in the decoder feature reconstruction stage to adapt and focus on related fields, ensuring the model's adaptation to different segmentation target features, and achieving different weights for the output of different convolution kernels to improve segmentation performance. We comprehensively evaluate our method on medical image segmentation tasks, by using four public datasets across CT and MRI. Remarkably, MFHARFNet method consistently outperforms other state-of-the-art methods, exceeding UNet by 2.1%, 0.9%, 6.6% and 1.0% on Dice on ATLAS, LiTs, BraTs2019 and Spine and intervertebral disc datasets, respectively. In addition, MFHARFNet minimizes network parameters and computational complexity as much as possible. The source codes are in https://github.com/OneHundred99/MFHARFNet.

An Improved Adaptive Grid-Based Progressive Triangulated Irregular Network Densification Algorithm for Filtering Airborne LiDAR Data

Ground filtering is crucial for airborne Light Detection and Ranging (LiDAR) data post-processing. The progressive triangulated irregular network densification (PTD) algorithm and its variants outperform others in accuracy, stability, and robustness, using grid-based seed point selection, TIN construction, and iterative rules for ground point identification. However, these methods still face limitations in removing low points and accurately preserving terrain details, primarily due to their sensitivity to grid size. To overcome this issue, a novel PTD filtering algorithm based on an adaptive grid (AGPTD) was proposed. The main contributions of the proposed method include an outlier removal method using a radius outlier removal algorithm and Kd-tree, a method for establishing an adaptive two-level grid based on point cloud density and terrain slope, and an adaptive selection method for angle and distance thresholds in the iterative densification processing. The performance of the AGPTD algorithm was assessed based on widely used benchmark datasets. Results show that the AGPTD algorithm outperforms the classical PTD algorithm in retaining ground feature points, especially in reducing Type I error and average total error significantly. In comparison with other advanced algorithms developed in recent years, the novel algorithm showed the lowest average Type I error, the minimal average total error, and the greatest average Kappa coefficient, which were 1.11%, 2.28%, and 90.86%, respectively. Additionally, the average accuracy, precision, and recall of AGPTD were 97.69%, 97.52%, and 98.98%, respectively.

Ultratough, Robustness, and Reprocessable Thermoset Epoxy Resins Synergistically Enhanced by Hierarchical Coordination Interactions and Dynamic Covalent Networks

AbstractEpoxy resins, ubiquitous and indispensable thermosetting material in various industries, suffer from the resistance to crack initiation, poor ductility, and irreparable permanent cross‐linked network, which hinder their utilization in high‐performance applications and are detrimental to the development of a sustainable economy. However, achieving the trade‐off between toughness, robustness, and reprocessability of epoxy resins remains a formidable challenge. Herein, an epoxy resin that combines ultratoughness, robustness, and reprocessability by incorporating a hierarchical crosslink structure embedded in a transesterification network is reported. The construction of hierarchical coordination structures facilitates formation of dense nano‐hard domains, enabling enhancement of both strength and toughness at multilength scales. As a result, the epoxy resin, integrating covalent adaptable networks with sacrificial non‐covalent networks, exhibits exceptional elongation at break (280%), modulus (1.8 GPa), stress at break (56.3 MPa), and tensile toughness (142 MJ m−3), showcasing its excellent endurance and ability to undergo multiple reprocessability. The application of this epoxy resin in carbon fiber‐reinforced polymer composites further underscores its potential as the resulting HCEV‐CFRPs exhibit unprecedented tensile properties and facilitate multiple non‐destructive recycling of high‐value carbon fiber under mild conditions. This research provides novel design principles for recyclable and high‐performance materials that combine strength and toughness.

Event-triggered finite-time adaptive neural network control for quadrotor UAV with input saturation and tracking error constraints

In this article, an event-triggered finite-time adaptive neural network control tracking strategy is proposed for quadrotor unmanned aerial vehicle (UAV) with input saturation and error constraints. Firstly, the radial basis function neural networks (RBFNNs) are adopted to identify the unknown uncertainty of quadrotor UAV model from the installation errors, gyroscope errors and so on. An auxiliary equation is constructed to deal with input physical saturation from the actuator motors. Additionally, by combining the performance function and error transformation, the issue of error constraint is solved. Based on the Lyapunov stability theory and event-triggered mechanisms, a finite-time adaptive neural network scheme is developed to ensure that the closed-loop quadrotor UAV system is semi-globally practically finite-time stable, and save the computation, resources, and transmission load. Finally, the simulation results illustrate the good tracking performance of quadrotor UAV by using the proposed control strategy.

Robust Self-Healing Adhesives Based on Dynamic Urethane Exchange Reactions.

Thermoset polyurethanes (PUs) have been successfully reprocessed as covalent adaptable networks (CANs) by catalyzing carbamate exchange. Here we extend bond exchange beyond the internal network cross-links to create a dynamic urethane adhesive. Interfacing PU CANs to substrates with nucleophilic functional groups creates adhesives capable of reversible transcarbamoylation with the substrate, which has not been demonstrated previously by CAN adhesives. Two types of thermoset PU films were synthesized, one containing the green carbamate exchange catalyst Zr(tmhd)4 and the other containing no catalyst. Although otherwise identical in chemical and network properties, as indicated by FT-IR spectroscopy and dynamic mechanical thermal analysis (DMTA), the film containing catalyst showed dynamic bond exchange behavior through stress relaxation analysis. When evaluated as an adhesive, the CAN film exhibited self-healing properties and retained its adhesive strength for five cycles, which is attributed to reversible covalent bonding to the glass substrate. This work expands industrially relevant CANs to structural adhesives and demonstrates their potential value in an application that presently employs PUs as single-use materials.

FCSSL: fusion enhanced contrastive self-supervised learning method for parallel MRI reconstruction

Objective. The implementation of deep learning in magnetic resonance imaging (MRI) has significantly advanced the reduction of data acquisition times. However, these techniques face substantial limitations in scenarios where acquiring fully sampled datasets is unfeasible or costly.Approach. To tackle this problem, we propose a fusion enhanced contrastive self-supervised learning (FCSSL) method for parallel MRI reconstruction, eliminating the need for fully sampledk-space training dataset and coil sensitivity maps. First, we introduce a strategy based on two pairs of re-undersampling masks within a contrastive learning framework, aimed at enhancing the representational capacity to achieve higher quality reconstruction. Subsequently, a novel adaptive fusion network, trained in a self-supervised learning manner, is designed to integrate the reconstruction results of the framework.Results. Experimental results on knee datasets under different sampling masks demonstrate that the proposed FCSSL achieves superior reconstruction performance compared to other self-supervised learning methods. Moreover,the performance of FCSSL approaches that of the supervised methods, especially under the 2DRU and RADU masks, but no need for fully sample data. The proposed FCSSL, trained under the 3× 1DRU and 2DRU masks, can effectively generalize to unseen 1D and 2D undersampling masks, respectively. For target domain data that exhibit significant differences from source domain data, the proposed model, fine-tuned with just a few dozen instances of undersampled data in the target domain, achieves reconstruction performance comparable to that achieved by the model trained with the entire set of undersampled data.Significance. The novel FCSSL model offers a viable solution for reconstructing high-quality MR images without needing fully sampled datasets, thereby overcoming a major hurdle in scenarios where acquiring fully sampled MR data is difficult.

Adaptive neural network backstepping control for the magnetorheological semi-active air suspension system with uncertain mass and time-varying input delay

Aiming at rejecting external disturbance induced by the random road profile, this paper proposes an adaptive robust control strategy to address the control problem of semi-active air suspension system installed with magnetorheological dampers (MRD) for enhancing ride comfort and handling performance. To effectively depict the inherent nonlinear dynamics of the adjustable air spring, a radial basis function neural network (RBFNN) model is trained by the experimental data offline, and an adaptive learning algorithm is introduced to maintain the modeling accuracy. Moreover, to handle the prevalent sensitive parameter variation (such as sprung mass), a nonlinear estimator is designed to estimate the uncertain sprung mass. Furthermore, a delay compensator is integrated into the control law to mitigate the impact of the time-varying input delay caused by the actuator of MRD to restrain the chattering phenomena. Additionally, the stability of the closed-loop system is rigorously proven by employing a Lyapunov–Krasovskii functional, guaranteeing the boundedness of both tracking error and estimation error. Simulation results are displayed and analyzed, illustrating the feasibility and efficiency of the proposed control strategy in depressing the sprung mass acceleration and tire deflection, showing its significance in both control performance enhancement and chattering elimination.

Retraction Note: Adaptive SDN-based network architecture for vehicle to vehicle communication using flexible optical networks for 5G

Retraction Note: Adaptive SDN-based network architecture for vehicle to vehicle communication using flexible optical networks for 5G

Finite-time robust control of morphing aircraft based on time-varying observer

Morphable vehicles have strong nonlinear, strong coupling and strong time-varying characteristics in the process of variation, which brings great challenges to the design of their flight control systems. To address this problem, a time-varying gain extended state observer is proposed to accurately approximate the system coupling terms. Combining adaptive backstepping control technology and finite-time control theory, a finite-time bounded robust adaptive controller is designed. By designing the barrier Lyapunov function, the actual control law and the adaptive network update law are obtained step by step, ensuring that the system command tracking error can converge to a predetermined range within a finite time. Finally, the effectiveness of the designed control system is verified by attitude control simulation. The simulation results show that the morphable vehicle can track the command signal well in the process of variation and is basically not affected by the variation rate.

Improving the estimation of the pile bearing capacity via hybridization technique based on adaptive network based fuzzy inference

Improving the estimation of the pile bearing capacity via hybridization technique based on adaptive network based fuzzy inference

Adaptive fusion graph convolutional network based interpretable fault diagnosis method for HVAC systems enhanced by unlabeled data

Fault diagnosis is critical in maintaining the operational stability and reliability of Heating, Ventilation, and Air Conditioning (HVAC) systems, which are crucial for ensuring indoor environmental quality and energy efficiency in buildings. However, traditional fault diagnosis methodologies face substantial challenges in accurately capturing the dynamic interactions within these systems and effectively utilizing unlabeled data. To overcome these limitations, an innovative fault diagnosis approach utilizing the Adaptive Fusion Graph Convolution Network (AFGCN) is proposed in this paper. This method significantly enhances the model's learning and inference abilities, particularly in scenarios with limited labeled data, by adaptively integrating the associative graph features of both unlabeled and labeled data. Furthermore, to augment the transparency and trustworthiness of the diagnostic outcomes, this paper introduces an interpretability analysis module. This module is designed to quantify the contribution of each sensor node in the fault diagnosis process. Experimental evaluations using the ASHRAE RP-1043, actual building operating chillers datasets, and LBNL FDD Data Sets_SDAHU indicate that this method offers substantial performance improvements in diagnosing faults in HVAC systems.

Supporting risk management through cyberspace: An adaptive network model simulating AI coach effects by inducing adherence to guidelines in neonatal medical protocols

In this article, it is shown how second-order adaptive agent-based network models can be used to support a medical team in healthcare institutions to adhere to specific Neonatal Hypoglycemia and Neonatal Hyperbilirubinemia treatment guidelines through the integration of an Artificial Intelligence (AI) Virtual Coach. The proposed AI Coach is designed to provide timely interventions and correct deviations when lapses in the health care practitioner’s internal mental model occur. Through simulating three different scenarios, the internal dynamics of these mental models, adaptive changes of these mental models (learning and forgetting), and the interaction between health care practitioners and the world is shown when: (1) There is perfect adherence to guidelines, (2) There is imperfect adherence to guidelines and (3) There is both perfect and imperfect adherence to guidelines alongside interventions of the AI Coach in the latter case.

Domain adaptive multi-frequency underwater image enhancement network

Domain adaptive multi-frequency underwater image enhancement network

Non-stationary mechanical sound source separation: An all-neural beamforming network driven by time–frequency convolution and self-attention

To tackle the challenging task of extracting target signal features amidst significant interference from complex, time-varying non-stationary noise in industrial scenarios, we propose a novel time-domain all-neural beamforming network tailored for non-stationary mechanical sound source separation, named TFANBNet. Leveraging time–frequency convolution and self-attention, TFANBNet promised robust performance in challenging acoustic environments. First, the proposed time–frequency convolution module employs parameterized complex-valued convolutional layers to simulate time–frequency transformations, thereby implicitly extracting time–frequency features. Then, the proposed adaptive attention transformer network enhances local attention interaction by integrating intermediate features, effectively enhancing long-range context modeling. Third, the beamforming module utilizes deep neural networks for beamforming weight estimation, replacing conventional noise covariance matrix computations, which have inherent performance limitations. Finally, the experimental results on the multi-channel spatial reverberation dataset synthesized from the MIMII dataset show that the proposed TFANBNet yields superior separation performance and generalization capabilities compared to the considered competitive methods.

Finite-time optimal control for a class of nonlinear systems with performance constraints via critic-only ADP: Theory and experiments

This paper addresses the optimal control problem within the framework of adaptive dynamic programming (ADP) for a class of nonlinear systems subjected to performance constraints. A new finite-time optimal control scheme is developed to stabilize the system by using the critic-only neural network ADP method. Compared with the existing ADP-based optimal control methods with uniformly ultimately bounded stability, the provided control scheme ensures that the controlled system's state and neural network weight estimation error are finite-time stable. It can ensure optimality, prescribed performance, and finite-time stability of the closed-loop control system simultaneously through an integration of ADP, the prescribed performance control technique, and Lyapunov theory. The designed adaptive neural network weight update law can relax the persisting excitation condition. The proposed control scheme is implemented on a robotic experiment platform to achieve trajectory tracking and verify its effectiveness.

Domain Transfer-Based Hypergraph Convolutional Network for Posture Anomaly Detection in Physical Education Teaching

Human skeleton-based posture anomaly detection has been widely applied in the field of physical education teaching. The existing spatio-temporal graph convolutional networks (ST-GCN) can fully utilize the local and global information of the human skeleton for action recognition, but the entire model requires a large amount of computation and the modeling of high-order relationships between joint points of the human skeleton is insufficient. To this end, this paper proposes a novel domain adaptive hypergraph convolutional network for basketball posture anomaly analysis by exploiting 2D skeleton information. First, we designed an effective hypergraph convolution feature extraction network to improve the high-order dependency modeling. To further improve the performance of the hypergraph convolutional network, we introduce domain adaptive learning technology to supervise the feature extraction learning of the target domain (2D skeleton) through the source domain (3D skeleton). At last, we construct a basketball action teaching analysis dataset for model evaluation. We conducted a large number of comparative experiments on the public dataset NTU RGB+D and our self-built dataset. All the results showed that our proposed hypergraph convolutional model effectively extracts features of 2D human skeletons, and by introducing domain adaptive learning, the performance of basketball anomaly detection is further improved.

Adaptive Cybersecurity Neural Networks: An Evolutionary Approach for Enhanced Attack Detection and Classification

The increasing sophistication and frequency of cyber threats necessitate the development of advanced techniques for detecting and mitigating attacks. This paper introduces a novel cybersecurity-focused Multi-Layer Perceptron (MLP) trainer that utilizes evolutionary computation methods, specifically tailored to improve the training process of neural networks in the cybersecurity domain. The proposed trainer dynamically optimizes the MLP’s weights and biases, enhancing its accuracy and robustness in defending against various attack vectors. To evaluate its effectiveness, the trainer was tested on five widely recognized security-related datasets: NSL-KDD, CICIDS2017, UNSW-NB15, Bot-IoT, and CSE-CIC-IDS2018. Its performance was compared with several state-of-the-art optimization algorithms, including Cybersecurity Chimp, CPO, ROA, WOA, MFO, WSO, SHIO, ZOA, DOA, and HHO. The results demonstrated that the proposed trainer consistently outperformed the other algorithms, achieving the lowest Mean Square Error (MSE) and highest classification accuracy across all datasets. Notably, the trainer reached a classification rate of 99.5% on the Bot-IoT dataset and 98.8% on the CSE-CIC-IDS2018 dataset, underscoring its effectiveness in detecting and classifying diverse cyber threats.

Beware of the Dark Side: The Rise of Cybercrimes

This study paper offers a top to bottom assessment of the mind boggling hardships introduced by cybercrime in India, pushed by the country's quick digitalization in different areas. The expansion in cybercrime occurrences, enveloping monetary extortion, information breaks, and cyberattacks, has arisen as a critical concern for partners in both public and business areas. This report investigations the essential drives executed by the Indian government, policing, and business elements to counter these difficulties. It inspects the developing complexities of digital dangers, evaluates the viability of current relief methods, and proposes thoughts for making a more adaptable and powerful network protection structure. The report offers fundamental bits of knowledge to upgrade India's safeguard against the raising scene of computerized dangers by tending to the essential causes and advancing nature of cybercrime in the country. The discoveries look to further develop India's online protection foundation and reinforce strength against expanding digital dangers.

Advanced security measures in coupled phase-shift STAR-RIS networks: A DRL approach

The rapid development of next-generation wireless networks has intensified the need for robust security measures, particularly in environments susceptible to eavesdropping. Simultaneous transmitting and reflecting reconfigurable intelligent surfaces (STAR-RIS) have emerged as a transformative technology, offering full-space coverage by manipulating electromagnetic wave propagation. However, the inherent flexibility of STAR-RIS introduces new vulnerabilities, making secure communication a significant challenge. To overcome these challenges, we propose a deep reinforcement learning (DRL) based secure and efficient beamforming optimization strategy, leveraging the deep deterministic policy gradient (DDPG) algorithm. By framing the problem as a Markov decision process (MDP), our approach enables the DDPG algorithm to learn optimal strategies for beamforming and transmission and reflection coefficients (TARCs) configurations. This method is specifically designed to optimize phase-shift coefficients within the STAR-RIS environment, effectively managing the coupled phase shifts and complex interactions that are critical for enhancing physical layer security (PLS). Through extensive simulations, we demonstrate that our DRL-based strategy not only outperforms traditional optimization techniques but also achieves real-time adaptive optimization, significantly improving both confidentiality and network efficiency. This research addresses key gaps in secure wireless communication and sets a new standard for future advancements in intelligent, adaptable network technologies.