Anomalization-based GRU and LSTM classifiers: An analyzation of intelligent intrusion detection

Rail trains are an indispensable part of the transportation, so technological innovations such as artificial intelligence will inevitably promote the overall development of unmanned driving technology in this field, how to ensure the safety of unmanned trains is related to intelligent rail detection technology is very important, such as the detection of foreign body intrusion on rails, turnouts, curved rails, etc. These technologies have become the basic technology of rail train automatic driving technology. This paper divides the key technology of rail intelligent detection into three parts, and conducts in-depth analysis and research from these three aspects: curved track detection and warning module, turnout detection and warning module, obstacle detection and warning module, and combined with the actual needs of each module, Continuously through experiments, choose more suitable and optimal algorithms to achieve and then integrate the module algorithms to achieve the synchronization of the work between each module without interfering with each other, and design and implement the intelligent rail detection and early warning system to enhance the interactive experience. According to the functional requirements of different modules of the system, this paper selects core algorithms such as curvature radius, Hough transforms and frame difference method to improve each module, and verifies the feasibility of the selected algorithm.KeywordsCurvature radiusHough probability transformationFrame difference method

In recent years, intelligent intrusion detection techniques based on machine learning have been the research spots in the field of intrusion detection. Whereas, as network traffic and network scale increase continually, some current machine learning algorithms can't meet the requirement of the network intrusion detection models for efficiency and accuracy, which restricts the application of machine learning into intrusion detection. In order to enhance the availability and practicality of intelligent intrusion detection system based on machine learning in high-speed network, an improved fast inductive learning method for intrusion detection (FILMID) is designed and implemented. Accordingly, an efficient intrusion detection model based on FILMID algorithm is presented. The experiment results on the standard testing dataset validate the effectiveness of the FILMID based intrusion detection model.

Malware authors are continuously developing crime toolkits. This has led to the situation of zero-day attacks, where malware harm computer systems despite the protection from existing Intrusion Detection Systems (IDSs). We propose an Intelligent Intrusion Detection and Prevention System (IIDPS) approach that combines the Signature based Intrusion Detection system (SIDS), Anomaly based Intrusion Detection System (AIDS) and Response Intrusion Detection System (RIDS). We used a risk assessment approach to determine an appropriate response action against each attack event. We also demonstrated the IIDPS make the detection and prevention of malware more effective.

Driven by recent technological advances in wireless communications, wireless sensors, and low power networked systems, wireless sensor networks (WSNs) are emerging as a promising technology in healthcare applications. Since information transmitted in these wireless body area networks (WBAN) often consists of critical and sensitive patient health and personal information, securing these networks is of central importance to their practical deployment in healthcare applications. The objective of this research is to design and develop intelligent intrusion detection techniques to improve security in WBAN. In this work, we propose iDetect, a multi-objective genetic algorithm based intrusion detection system (IDS) to provide optimal attack detection in these networks. The proposed algorithm guarantees that only those features necessary for detecting a specific attack is used in the intrusion detection process, thereby decreasing the computational complexity.

To study the characteristics and performance of the deep learning in intelligent intrusion detection, two hybrid algorithms, which combine restricted Boltzmann machine (RBM) with support vector machine (SVM) and deep belief network (DBN) respectively, are used to analyze the accuracy, false positive rate, false negative rate and testing time with the data set used for The Third International Knowledge Discovery and Data Mining Tools Competition (KDDCup99). Compared with each other and traditional hybrid intrusion detection algorithm, DBN performs better than the other both in the accuracy and speed, which is attributed to the unsupervised learning of RBM networks and the combination of the neural networks at the bottom.

The widespread use of the Internet of Things and distributed heterogeneous devices has shed light on the implementation of efficient and reliable intrusion detection systems. These systems should be able to efficiently protect data and physical devices from cyber-attacks. However, the huge amount of data with different dimensions and security features can affect the detection accuracy and increase the computation complexity of these systems. Lately, Artificial Intelligence has received significant interest and is now being integrated into these systems to intelligently detect and protect against cyber-attacks. This paper aims to propose an intelligent intrusion detection model to predict and detect attacks in cyberspace. The model is designed based on the concept of Decision Trees, taking into consideration the ranking of the security features. The model is applied to a real dataset for network intrusion detection systems. Moreover, it is validated based on predefined performance evaluation metrics, namely accuracy, precision, recall and Fscore. Meanwhile, the experimental results reveal that our tree-based intrusion detection model can detect and predict cyber-attacks efficiently and reduce the complexity of computation process compared to other traditional machine learning techniques.

Blockchain and federated learning-based intrusion detection approaches for edge-enabled industrial IoT networks: a survey

An intelligent community intrusion detection system based on wireless sensor network and fuzzy neural network (FNN) is presented in this paper. This system is composed of ARM (Advanced RISC Machines) data acquisition nodes, wireless mesh network and control centre. The data acquisition node uses sensors to collect information and processes them by image detection algorithm, and then transmits information to control centre with wireless mesh network. When there is abnormal phenomenon, the system starts the camera and the FNN is used to recognize the face image. We construct the structure of fuzzy neural network, and adopt the Levenberg-Marquart optimizing algorithm to train fuzzy neural network, thereby enhancing the convergence rate and the recognition accuracy. With the ability of strong self-learning and nonlinear function approach of fuzzy neural network, the recognition method can truly classify the face. This system resolves the defect and improves the intelligence and alleviates worker's working stress.

A flexible intrusion detection and response system (ID&R) needs to maximize security while minimizing cost and making response automatically. CI/sup 2/D&R, the cost-based intelligent intrusion detection and response system, is proposed, which is originally developed as a facility to deal with network-based attacks and to make effective response automatically and intelligently. The networking environment deployed with the CI/sup 2/D&R consists of two major parts: guard, which runs on the specific guarded host (GH), and spy, which runs in guarded network (GN). The components of the CI/sup 2/D&R are introduced, which include intrusion detection, attack classification, damage analysis, attack path rebuilding, resources automatically safeguarding, calamity recovery, and security officer. The several kinds of data flow in CI/sup 2/D&R are discussed, too. While CI/sup 2/&R is only a prototype, some experimental results are also presented.

The health of a computer network needs to be assessed and protected in much the same manner as the health of a person. The task of an intrusion detection system is to protect a computer system by detecting and diagnosing attempted breaches of the integrity of the system. A robust intrusion detection system for a computer network will necessarily use multiple sensors, each providing different types of information about some aspect of the monitored system. In addition, the sensor data will often be analyzed in several different ways. We describe a decision engine for an intelligent intrusion detection system that fuses information from different intrusion detection modules using a causal knowledge based inference technique. Fuzzy cognitive maps (FCMs) and fuzzy rule-bases are used for the causal knowledge acquisition and to support the causal knowledge reasoning process.

Intelligent intrusion detection system featuring a virtual fence, active intruder detection, classification, tracking, and action recognition

An Intelligent Intrusion Detection and Mitigation Model Using Self-Supervised Auto Encoders and Ensemble Learning

Epidemic malicious codes including Internet worms and botnets have continuously evolved to be more intelligent and complicated. In particular, the recent distributed denial-of-service (DDoS) attack that occurred in United States and South Korea in July, 2009 gives an opportunity to reconsider the epidemic malicious code. Since automatic patching systems and intelligent intrusion detection and prevention systems mitigate rapid infection, fast infections such as Slammer-like worms cannot successfully spread. As of the 2009 July DDoS attack, malicious codes prefer hiding their malicious activities and trying to infect others silently until D-day. Since slow infection is difficult to detect by the current IDS or IPS, this infection strategy is likely to become prevalent. In a slow infection, the incubation period is a key factor in determining the extent to which an epidemic malicious code spreads. This study provides an analysis framework to understand the impact of incubation period in the spread of epidemic malicious code. Intuitively, a longer latent period increases the number of infected hosts, but the detection probability also increases. This trade-off suggests an optimal incubation period determination problem to maximize the number of infected hosts. Solving this problem is essential to predicting the explicit or implicit intention of attackers and to counteract against the attack in a strategic manner. Through analysis and simulations, we provide data and insight regarding epidemic malicious code that exploits incubation period.

This paper presents an adaptive deep neural network (DNN) approach for intrusion detection and prevention in shipboard AC/DC microgrids, focusing on shore-to-ship power connections and power quality (PQ) challenges. The method addresses false data injection attacks (FDIAs) that disrupt secondary control, leading to voltage and current regulation failures. By integrating Fast Fourier Transform (FFT) for frequency-domain feature extractions such as voltage sag/swell, harmonics, and transient distortions with DNN classification, the model achieves high accuracy in detecting cyberattacks under PQ disturbances. Optimized using the Adam optimizer and ReLU-sigmoid activation, the framework enhances detection accuracy while reducing false positives. The obtained results demonstrate superior performance of FFT-DNN over state-of-the-art methods, achieving 97.7% accuracy across attack scenarios. The approach effectively classifies between cyber-intrusions and power quality disturbances, offering a scalable cybersecurity solution for maritime systems. This study advances secure and resilient shore-ship DC microgrids by addressing cybersecurity vulnerabilities and power quality challenges in shore-power connections.

With the increasing scale of network, attacks against network emerge one after another, and security problems become increasingly prominent. Network intrusion detection system is a widely used and effective security means at present. In addition, with the development of machine learning technology, various intelligent intrusion detection algorithms also start to sprout. By flexibly combining these intelligent methods with intrusion detection technology, the comprehensive performance of intrusion detection can be improved, but the vulnerability of machine learning model in the adversarial environment can not be ignored. In this paper, we study the defense problem of network intrusion detection system against adversarial samples. More specifically, we design a defense algorithm for NIDS against adversarial samples by using bidirectional generative adversarial network. The generator learns the data distribution of normal samples during training, which is an implicit model reflecting the normal data distribution. After training, the adversarial sample detection module calculates the reconstruction error and the discriminator matching error of sample. Then, the adversarial samples are removed, which improves the robustness and accuracy of NIDS in the adversarial environment.