The relationship between trust and consensus has long been a focal point in social network group decision-making (SN-GDM). This article investigates a two-stage feedback model for SN-GDM driven by a trust incentive mechanism, aimed at promoting expert collaboration and enhancing consensus levels. First, a trust path determination method is proposed by integrating trust score and social authority to identify the optimal trust propagation paths. Then, based on trust utility (TU), an improved Uninorm trust propagation operator is developed to better align with real-world SN-GDM scenarios. Next, a trust incentive mechanism is designed, incorporating cooperative behavior into a dynamic social trust network to construct a two-stage feedback model. The first stage is a minimum-cost consensus model, and the second is a maximum-consensus model, both constrained by the cooperation degree index (CDI). A social trust network correction mechanism is further introduced, with results showing that increased consensus enhances mutual trust among experts, thereby improving CDI values. Consequently, the proposed trust-incentive-driven two-stage feedback model effectively facilitates consensus building within expert groups. Compared with traditional feedback models, this approach better reflects the psychological dynamics of experts during the consensus-reaching process and further explores its applicability in practical decision-making contexts. Finally, the proposed model is applied to a real-world decision scenario and compared with other methods, demonstrating its effectiveness and applicability.

In the consensus reaching process (CRP) of social network group decision making (SN-GDM), the non-cooperative behavior exhibited by experts will hinder the achievement of group consensus. This paper develops a non-cooperative behavior management framework based on trust propagation and dynamic cooperation index under bidirectional feedback context. On the one hand, a trust propagation operator with trust decay is established to enhance the trust relationship between non-cooperative experts; On the other hand, the fuzzy preference relations are utilized as preference expression structure, and the mutual reinforcing effect between consensus and trust is explored to achieve the dynamic enhancement of cooperation index, thereby facilitating the transformation of non-cooperative behavior. Specifically, a cooperation index is formulated to identify the non-cooperation behavior. Subsequently, a non-cooperative behavior transformation method by dynamic cooperation index is investigated. Finally, a bidirectional feedback mechanism is provided for group consensus reaching. This paper provides an innovative strategy for detecting and managing non-cooperative behavior, an illustrative example and some analyses are presented to verify the validity of proposed method.

Multipath interference in radar signals caused by sea, ground, and other environments poses significant challenges to the target detection, tracking, and classification capabilities of radar systems. Existing methods for radar signal identification require labeled samples and focus mainly on the classification of normal signals. However, in practice, anomalous samples (multipath interference signals) may be scarce and highly imbalanced (i.e., mostly normal samples). To address this problem, we propose a deep anomaly detection with attention (DADA) for semisupervised detection of multipath radar signals. The method transforms radar signals into time–frequency images and is trained exclusively on normal samples. The autoencoder architecture is extended with a feature extractor network to capture latent sample features. CBAM attention is introduced to improve feature extraction. By learning the distribution of normal samples in high‐dimensional image space and low‐dimensional feature space, a two‐dimensional feature space representing normal samples is constructed. A one‐class SVM then learns the boundary of normal samples for anomaly detection. Extensive experiments on radar signal datasets validate the effectiveness of the proposed approach.

In the intelligent high-speed railway system, the security of deep neural networks-based high-speed train bogie fault diagnosis methods is challenged by adversarial attacks, which can mislead the model predictions with maliciously designed adversarial examples. However, existing methods do not consider the robustness against adversarial attacks. To address the aforementioned challenge, we propose a novel method called AdvSifter to perform robust fault diagnosis for the high-speed train bogie against adversarial attacks, which leverages adversarial training (AT) to guarantee the model with fundamental adversarial robustness. Besides, a defense algorithm called residual perturbation inversion (RPI) is developed to recover and remove the perturbations in adversarial examples to reduce the power of the adversarial examples. A defense module called SifterNet is designed to perform RPI to further improve the adversarial robustness of AdvSifter on the base of AT. Experimental results on a high-speed train bogie monitoring dataset demonstrate that our method outperforms state-of-the-art methods by a large margin.

Quantifying the abnormal degree of each instance within data sets to detect outlying instances, is an issue in unsupervised anomaly detection research. In this paper, we propose a robust anomaly detection method based on principal component analysis (PCA). Traditional PCA-based detection algorithms commonly obtain a high false alarm for the outliers. The main reason is that ignores the difference of location and scale to each component of the outlier score, this leads to the cumulated outlier score deviates from the true values. To address the issue, we introduce the median and the Median Absolute Deviation (MAD) to rescale each outlier score that mapped onto the corresponding principal direction. And then, the true outlier scores of instances can be obtained as the sum of weighted squares of the rescaled scores. Also, the issue that the assignment of the weight for each outlier score will be solved. The main advantage of our new approach is easy to build with unsupervised data and the recognition performance is better than the classical PCA-based methods. We compare our method to the five different anomaly detection techniques, including two traditional PCA-based methods, in our experiment analysis. The experimental results show that the proposed method has a good performance for effectiveness, efficiency, and robustness.

We propose a new multiscale rotation-invariant convolutional neural network (MRCNN) model for classifying various lung tissue types on high-resolution computed tomography. MRCNN employs Gabor-local binary pattern that introduces a good property in image analysis-invariance to image scales and rotations. In addition, we offer an approach to deal with the problems caused by imbalanced number of samples between different classes in most of the existing works, accomplished by changing the overlapping size between the adjacent patches. Experimental results on a public interstitial lung disease database show a superior performance of the proposed method to state of the art.

In this paper, we present a new approach to count the number of people that cross a counting line from video images. This paper focuses on point-level annotation in training images and incorporate spatial features along with novel temporal features in training the structured random forest for estimating crowd density. By computing the crowd velocity, we model the crowd counting map as elementwise multiplication of crowd density map and crowd velocity map. Integrating over crowd counting map on the line of interest(LOI) locations leads to the instantaneous LOI counting numbers. We show that results are comparable to those obtained when using more complex and costly techniques.

Deep Belief Network (DBN) learns the features of the raw data automatically, and develops a new idea for the study of fault analysis of High Speed Train (HST). Combining deep learning and classification ensemble technology, this paper presents a novel DBN hierarchical ensemble model for HST fault analysis. Firstly, Fast Fourier Transform (FFT) coefficients of the HST vibration signals are extracted as the state of the visible layer of the model, and then DBN is used to learn the hierarchical features of the vibration signals automatically. The features of each layer learned by DBN are used to train Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Radial Basis Function (RBF) Neural Network respectively. Finally, the Majority Voting (MV), the Classification Entropy Voting Principle (CE), and the Winner Takes All (WTA) ensemble strategies are used for combination to get the final results. The experiments are conducted by using laboratory data sets and simulation data sets. The results show that the fault recognition rate of the proposed model is much higher than the traditional fault analysis methods. In addition, unlike the DBN model, the proposed model is affected slightly by the number of network layers and the size of hidden units.

Feature extraction is one of key steps in fault diagnosis for High Speed Train (HST). In this work, we present a method that can automatically extract high-level features from HST vibration signals and recognize the faults. The method is composed of a Deep Belief Network (DBN) on Fast Fourier Transform (FFT) of vibration signals. DBNs can be trained greedily, layer by layer, using a model referred to as a Restricted Boltzmann Machine (RBM). The real data sets and simulation data sets of HST vibration signals are selected in experiments. First, the vibration signals are preprocessed by FFT. Then, the FFT coefficient-vectors are used to set the states of the visible units of DBNs. Finally, n label units are connected to the "top" layer of the DBNs to identify different faults. The experimental results show that the method may learn useful high-level features from vibration signals and diagnose the different faults of HST.

A novel complex-valued non-linear equaliser-based pipelined decision feedback recurrent neural network (CPDFRNN) is proposed in this study for non-linear channel equalisation in wireless communication systems. The CPDFRNN with low computational complexity, a modular structure comprising a number of modules that are interconnected in a chained form, is an extension of the recently proposed real-valued pipelined decision feedback recurrent neural equalisers. Each module is implemented by a small-scale complex-valued decision feedback recurrent neural network (CDFRNN). Moreover, a decision feedback part in each module can overcome the unstable characteristic of the complex-valued recurrent neural network (CRNN). To suit the modularity of the CPDFRNN, an adaptive amplitude complex-valued real-time recurrent learning (CRTRL) algorithm is presented. Simulations demonstrate that the CPDFRNN equaliser using the amplitude CRTRL algorithm with less computational complexity not only eliminates the adverse effects of the nesting architecture, but also provides a superior performance over the CRNN and CDFRNN equalisers for non-linear channels in wireless communication systems.

This correspondence presents an extended pipelined second-order Volterra (EPSOV) filter for active control of nonlinear noise processes. The corresponding nonlinear filtered-x algorithms using the filter bank implementation are also suggested. Compared to the standard SOV filter using the filtered-x least mean square (SOVFXLMS), those modules of the EPSOV filter can be performed simultaneously in a pipelined parallelism fashion, and this would lead to a significant improvement in its total computational efficiency. Results obtained from computer simulations for nonlinear noise processes demonstrate that the proposed method outperforms the SOV.

In this paper, a novel approach based on Gaussian Chirplet Atoms is presented to automatically recognise radar emitter signals. Firstly, based on the over-completed dictionary of Gaussian Chirplet atoms, the improved matching pursuit (MP) algorithm is applied to extract the features of the time-frequency atoms from the typical radar emitter signals, and FFT is introduced to effectively reduce the time complexity of searching step of MP. Secondly, reduce dimension of the feature parameters to re-extract the classification feature vectors. Finally, adopt the hierarchy decision strategy to realize automatic classification. The simulation experiment result shows that the classification feature vector has good properties of clustering the same and separating the different kind of radar emitter signals. Over 90% recognition accuracy can be achieved as the signal-to-noise ratio is greater than -4dB. Therefore, the approach of signal recognition is feasible in the practical engineering area.

Security auditing and worm containment is used to guarantee the network security in metropolitan area networks. Multi-agent system for security auditing and worm containment in MAN (MSAWCM) is presented to audit user's accesses and provide a first-class automatic reaction mechanism that automatically applies containment strategies to prevent clean host from being infected by blocking the propagation of the worms MSAWCM uses broadband access server as information gathering agent that uses hardware packet filter (HPF) to get packet from MAN. It adaptively studies and audits the accessing in the whole network and dynamically changes the working parameters to detect the unknown worms. MSAWCM integrates worm detection system (WDS) and network management system (NMS). Reaction measures can be taken by using SNMP interface to control BAS as soon as the WDS detect the active worm. MSAWCM is very effective in blocking random scanning worms that are very noisy and tend to waste a lot of network bandwidth and crash routers. Simulation results indicate that high worm infection rate of epidemics can be avoided to a degree by MSAWCM blocking the propagation of the worms.

An important study in multi-agent systems is the development of cooperative behavior between agents that have a shared goal. In this paper, an example of the predator-prey problem is studied in which four predator agents, using the reinforcement learning method, in an attempt to collectively achieve the task of surrounding one prey agent. First, we describe the structure of the predator agents and the prey agent with their state sensing capability, the action learning method and the action choosing mechanism. Next we study two cooperative behavior mechanisms between agents of multi-agent systems in predator-prey problem. Finally, we demonstrate that cooperative agents outperform agents without cooperative behavior according to simulations of the predator-prey problem and display the corresponding experimental results.

Radar emitter signal recognition plays an important role in electronic intelligence systems and electronic support measure systems. To heighten accurate recognition rate of radar emitter signals, this paper proposes a hierarchical classifier structure to recognize radar emitter signals. The proposed structure combines resemblance coefficient classifier, support vector machines with binary tree architecture and linear classifier based on Mahalanobis distance. Experimental results of recognizing multiple radar emitter signals show that the introduced classifier is simpler, consumes smaller training time and achieves higher accurate recognition rate and greater efficiency, in comparison with one-versus-rest support vector machines, one-versus-one support vector machines and binary-tree support vector machines.