Enhancing Network Intrusion Detection Performance using Generative Adversarial Networks to Generate Synthetic Attack Samples
Leveraging the power of Generative Adversarial Networks (GANs) to generate synthetic network traffic data that closely mimics real-world anomalous behavior, in order to enhance the performance of network intrusion detection systems (NIDS) by addressing the challenge of limited training data for attack samples.
Enhancing Trustworthiness of Machine Learning-Based Network Intrusion Detection Systems through Uncertainty Quantification
Proper uncertainty quantification is crucial for developing trustworthy machine learning-based intrusion detection systems that can reliably detect known attacks and identify unknown network traffic patterns.
Exploiting Vulnerabilities in NAT Strategies and Router Behaviors to Hijack TCP Connections in Wi-Fi Networks
An off-path attacker can exploit vulnerabilities in the NAT port preservation strategy and insufficient reverse path validation of Wi-Fi routers to infer active TCP connections, evict the original NAT mappings, and reconstruct new mappings to intercept the sequence and acknowledgment numbers, enabling them to terminate, hijack, or inject traffic into the victim's TCP connections.
Continuous Training Reduces the Effectiveness of Adversarial Attacks Against Machine Learning-based Network Intrusion Detection Systems
Continuous retraining of machine learning models, even without adversarial training, can significantly reduce the effectiveness of adversarial attacks against network intrusion detection systems.
General In-Network Unsupervised Intrusion Detection by Rule Extraction
Genos, a general in-network framework for unsupervised anomaly-based network intrusion detection, achieves high throughput, interpretability, and trivial updating overhead by extracting model-agnostic rules.
Addressing Class Imbalance in Bot-IoT Dataset for Effective Network Intrusion Detection
To address the class imbalance problem in the Bot-IoT dataset, a binary classification method with synthetic minority over-sampling techniques (SMOTE) is proposed to effectively detect attack packets in IoT network traffic.
4G/5G Low-Layer Control Procedures Vulnerabilities Exposed
Lower layers of 4G/5G networks are vulnerable to passive and active attacks due to lack of encryption and integrity protection.
4G/5G Low-Layer Control Procedures Vulnerabilities Exposed
4G/5G low-layer control procedures are vulnerable to passive and active attacks, leading to user tracking, communication disruption, and privacy breaches.
Superflows: A New Tool for Forensic Network Flow Analysis
Superflows propose a new method for grouping network flows based on common hypotheses to enhance operational network response.
ML-based Malicious Traffic Detection for Terabit Networks: Peregrine
Peregrine improves detection performance by offloading feature computation to the network data plane, enhancing efficiency and scalability for Terabit networks.
© 2024 by Linnk AI