Cascading Bagging and Boosting Ensemble Methods for Intrusion Detection in Cyber‐Physical Systems

Security and Privacy, Journal Year: 2025, Volume and Issue: 8(1)

Published: Jan. 1, 2025

ABSTRACT An ensemble‐inspired intrusion detection approach for cyber‐physical systems (CPSs) has been proposed. CPSs are susceptible to different cyber‐attacks. Attacks on result in the hindrance of critical services made available by them. To protect various (IDSs) exist. However, present IDSs have limitations constrained accuracy, high false alarm rate, and latency. A hybrid that integrates pros bagging boosting methods proposed minimize mentioned limitations. The uses AdaBoost random forest (RF) algorithms as base models. Optimal features most indicative attack behavior selected based aggregated significant scores each feature calculated using models retrained features. final selection is predicted adopting majority vote technique. method implemented CIC‐IoT‐2023 dataset multiclassification intrusions. Thereby selecting only best‐exclusive number detection, gives better results like accuracy (98.27%), precision (0.98), recall F1‐score average positive rate (FPR) (0.0006), testing time (0.1563s). By conducting extensive experiments, it observed best 21 out 46 aids minimizing space complexities approach. performs than existing state‐of‐the‐art approaches literature regarding FPR time.

Language: Английский

---

Securing the 6G–IoT Environment: A Framework for Enhancing Transparency in Artificial Intelligence Decision-Making Through Explainable Artificial Intelligence

Sensors, Journal Year: 2025, Volume and Issue: 25(3), P. 854 - 854

Published: Jan. 30, 2025

Wireless communication advancements have significantly improved connectivity and user experience with each generation. The recent release of the framework M.2160 for upcoming sixth generation (6G or IMT-2030) cellular wireless standard by ITU-R has heightened expectations, particularly Internet Things (IoT) driven use cases. However, this progress introduces significant security risks, as technologies like O-RAN, terahertz communication, native AI pose threats such eavesdropping, supply chain vulnerabilities, model poisoning, adversarial attacks. increased exposure sensitive data in 6G applications further intensifies these challenges. This necessitates a concerted effort from stakeholders including ITU-R, 3GPP, ETSI, OEMs researchers to embed resilience core components 6G. While research is advancing, establishing comprehensive remains challenge. To address evolving threats, our proposes dynamic that emphasizes integration explainable (XAI) techniques SHAP LIME advanced machine learning models enhance decision-making transparency, improve complex environments, ensure effective detection mitigation emerging cyber threats. By refining accuracy ensuring alignment through recursive feature elimination consistent cross-validation, approach strengthens overall posture IoT–6G ecosystem, making it more resilient attacks other vulnerabilities.

Language: Английский

---

Internet of Things (IoT) Technologies in Cybersecurity: Challenges and Opportunities

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(6), P. 2935 - 2935

Published: March 8, 2025

The continuous development and increasing availability of Internet Things (IoT) solutions have led to an era connectivity in which everyday objects—from household appliances industrial machines—are connected via the [...]

Language: Английский

---

Optimized Intrusion Detection Approach for Cyber‐Physical System Using Meta‐Learning With Stacked Generalization: An Ensemble Learning Inspired Approach

Security and Privacy, Journal Year: 2025, Volume and Issue: 8(3)

Published: April 27, 2025

ABSTRACT Cyber‐physical systems (CPSs) are crucial in providing vital infrastructure like smart grids, cities, automobiles, healthcare systems, and so forth, for many nations. CPSs vulnerable to various attacks due their large attack surface. An on these may lead the disruption of critical services. To protect an optimized intrusion detection approach is needed. Although approaches exist, they have limitations poor accuracy, high time, space time complexities, false alarm rates, etc. stack generalized meta‐learner‐based has been proposed this paper. The utilizes numerous core models a meta‐learner classify network traffic CPSs. base trained learning data, outcomes used as input features meta‐learner, which then makes final prediction. Four classifiers being models, namely random forest (RF), gradient boosting (GB), multiple layer perceptron (MLP), k ‐nearest neighbors (KNNs), extreme (XGB) classifier meta‐learner. predictions generated using stacking ensemble approach. Auto encoders feature extraction, thereby utilizing unique objective function designed recursive attribute elimination. presented selects only 10 out 46 features, helps reducing complexities. While implementing CIC‐IoT‐2023 dataset, following results obtained: multi‐classification accuracy (98.94%), precision (0.99), recall F 1 score average positive rate (0.0003), (0.12 s). When implemented NSL‐KDD (99%), (0.0012). UNSW‐NB15 (99.56%), (0.0002). performs better contrast other cutting‐edge approaches. Also, introduces novel effective strategy

Language: Английский

---

A secure and trustworthy blockchain-assisted edge computing architecture for industrial internet of things

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: May 2, 2025

Language: Английский

---

Collaborative intrusion detection using weighted ensemble averaging deep neural network for coordinated attack detection in heterogeneous network

International Journal of Information Security, Journal Year: 2024, Volume and Issue: 23(5), P. 3329 - 3349

Published: July 23, 2024

Abstract Detecting coordinated attacks in cybersecurity is challenging due to their sophisticated and distributed nature, making traditional Intrusion Detection Systems often ineffective, especially heterogeneous networks with diverse devices systems. This research introduces a novel Collaborative System (CIDS) using Weighted Ensemble Averaging Deep Neural Network (WEA-DNN) designed detect such attacks. The WEA-DNN combines deep learning techniques ensemble methods enhance detection capabilities by integrating multiple (DNN) models, each trained on different data subsets varying architectures. Differential Evolution optimizes the model’s contributions calculating optimal weights, allowing system collaboratively analyze network traffic from sources. Extensive experiments real-world datasets like CICIDS2017, CSE-CICIDS2018, CICToNIoT, CICBotIoT show that CIDS framework achieves an average accuracy of 93.8%, precision 78.6%, recall 60.4%, F1-score 62.4%, surpassing models matching performance local DNN models. demonstrates practical benefits improving environments, offering superior adaptability robustness handling complex attack patterns.

Language: Английский

---

CNN-GWO-voting & hybrid: ensemble learning inspired intrusion detection approaches for cyber-physical systems

DELETED, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 26, 2024

Cyber threats are increasingly dynamic and sophisticated, often surpassing the capabilities of conventional intrusion detection systems (IDSs). Current IDSs for cyber-physical (CPSs) face limitations such as high computational complexity, low accuracy, elevated false positive rates. To address these challenges, two novel IDS approaches inspired by ensemble learning being proposed. The first approach, CNN-GWO-Voting, combines deep learning, evolutionary optimization, to enhance effectiveness. This approach introduces a hybrid model design with soft voting mechanism fitness function optimize attribute selection, distinguishing it from existing models. In this convolutional neural network (CNN) is used feature extraction, gray-wolf optimizer (GWO) selecting optimal attack relevant features, blend predictions four base classifiers: random forest, support vector machine (SVM), decision tree XGBoost. Evaluated on CIC-IoT-2023 dataset, achieves accuracy (99.15%), precision (0.99), recall F1-score FPR (0.008), FNR (0.008) while only 15 46 significantly reducing complexity. second comprehensive classifier binary classification, utilizing logistic regression, naïve bayes, SVM, k-nearest neighbour, multilayer perceptron classifiers, technique voting, stacking, bagging, boosting. On boosting yielded best results i.e. (98.16%), (0.98), (0.98). Both proposed not outperform recent advanced techniques but also introduce significant improvements in methodologies tailored specifically CPSs security.

Language: Английский

---

Hybrid Enhanced Intrusion Detection Frameworks for Cyber-Physical Systems via Optimal Features Selection

Indian Journal of Science and Technology, Journal Year: 2024, Volume and Issue: 17(30), P. 3069 - 3079

Published: Aug. 2, 2024

Background/Objectives: Cyber-physical systems (CPSs) form the critical infrastructure for many nations like smart grids, home automation, cities, health care, automobiles, etc. These are susceptible to various attacks due their wider surface area. Cyber-attacks on these can interrupt services provided by them. Thus, intrusion detection frameworks (IDFs) needed identify CPSs so that countermeasures be taken minimize harm of such attacks. Limitations existing IDFs poor rate, high time, false alarm and large space time complexities. The objective this study is design hybrid-enhanced overcome issues. Methods: Two enhanced proposed in research work. SelectKBest-MI (mutual information), framework fuses two filter-based feature selection techniques namely SelectKBest mutual information selecting optimal features, Random Forest (RF) utilized as a classifier. second IDF named CNN-SVM-GWO. Convolutional Neural Network (CNN) used extraction attributes, Support Vector Machine (SVM) Gray Wolf Optimizer (GWO) number selection, RF Extreme Gradient Boosting (XGB) classifiers detection. datasets have been used: CICIDS2017 CIC-IoT-2023. Parameters considered comparison with accuracy, precision, recall, F1-score, prediction time. Findings: Implementation using dataset, results better accuracy 99.99%, precision 0.99, recall F1-score 0.99 binary classification. CNN-SVM-GWO CIC-IoT-2023 dataset 99.60%(RF), 99.49(XGB), 0.99. Novelty: 0.75 seconds 0.078 (XGB). model has reduced complexity. Novel hybrid efficiency. Keywords: systems, Intrusion system, Optimal wolf optimizer, neural network

Language: Английский

---

Exploration of Trust-Based Collaborative Intrusion Detection System Development: A Survey

Published: July 13, 2024

Language: Английский

---

Taxonomy and Survey of Collaborative Intrusion Detection System using Federated Learning

ACM Computing Surveys, Journal Year: 2024, Volume and Issue: 57(4), P. 1 - 36

Published: Oct. 25, 2024

This review article looks at recent research on Federated Learning (FL) for Collaborative Intrusion Detection Systems (CIDS) to establish a taxonomy and survey. The motivation behind this comes from the difficulty of detecting coordinated cyberattacks in large-scale distributed networks. anomalies are one network that need be detected through robust collaborative learning methods. FL is promising method research. aims offer insights lesson learn creating anomaly detection CIDS using as method. Our findings suggest required map discussion area, including an algorithm training model, dataset, global aggregation system architecture, security, privacy. results indicate approach CIDS, proposed could useful future area. Overall, contributes growing knowledge providing lessons researchers practitioners. also concludes significant challenges, opportunities, directions based FL.

Language: Английский

---