Blockchain security enhancement: an approach towards hybrid consensus algorithms and machine learning techniques

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Jan. 11, 2024

Abstract In this paper, we propose hybrid consensus algorithms that combine machine learning (ML) techniques to address the challenges and vulnerabilities in blockchain networks. Consensus Protocols make ensuring agreement among applicants distributed systems difficult. However, existing mechanisms are more vulnerable cyber-attacks. Previous studies extensively explore influence of cyber attacks highlight necessity for effective preventive measures. This research presents integration ML with proposed advantages over predicting cyber-attacks, anomaly detection, feature extraction. Our approaches leverage optimize protocols' security, trust, robustness. also explores various algorithms, such as Delegated Proof Stake Work (DPoSW), (PoSW), CASBFT (PoCASBFT), Byzantine (DBPoS) security enhancement intelligent decision making protocols. Here, demonstrate effectiveness methodology within decentralized networks using ProximaX platform. study shows framework is an energy-efficient mechanism maintains adapts dynamic conditions. It integrates privacy-enhancing features, robust mechanisms, detect prevent threats. Furthermore, practical implementation these ML-based models faces significant challenges, scalability, latency, throughput, resource requirements, potential adversarial attacks. These must be addressed ensure successful network real-world scenarios.

Language: Английский

---

MLSTL-WSN: machine learning-based intrusion detection using SMOTETomek in WSNs

International Journal of Information Security, Journal Year: 2024, Volume and Issue: 23(3), P. 2139 - 2158

Published: March 19, 2024

Abstract In the domain of cyber-physical systems, wireless sensor networks (WSNs) play a pivotal role as infrastructures, encompassing both stationary and mobile sensors. These sensors self-organize establish multi-hop connections for communication, collectively sensing, gathering, processing, transmitting data about their surroundings. Despite significance, WSNs face rapid detrimental attacks that can disrupt functionality. Existing intrusion detection methods encounter challenges such low rates, computational overhead, false alarms. issues stem from node resource constraints, redundancy, high correlation within network. To address these challenges, we propose an innovative approach integrates machine learning (ML) techniques with Synthetic Minority Oversampling Technique Tomek Link (SMOTE-TomekLink) algorithm. This blend synthesizes minority instances eliminates links, resulting in balanced dataset significantly enhances accuracy WSNs. Additionally, incorporate feature scaling through standardization to render input features consistent scalable, facilitating more precise training detection. counteract imbalanced WSN datasets, employ SMOTE-Tomek resampling technique, mitigating overfitting underfitting issues. Our comprehensive evaluation, using network (WSN-DS) containing 374,661 records, identifies optimal model The standout outcome our research is remarkable performance model. binary classification scenarios, it achieves rate 99.78%, multiclass attains exceptional 99.92%. findings underscore efficiency superiority proposal context detection, showcasing its effectiveness detecting intrusions

Language: Английский

---

A Comparative Study of Anomaly Detection Techniques for IoT Security Using Adaptive Machine Learning for IoT Threats

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 14719 - 14730

Published: Jan. 1, 2024

Anomaly detection is a critical aspect of various applications, including security, healthcare, and network monitoring. In this study, we introduce FusionNet, an innovative ensemble model that combines the strengths multiple machine learning algorithms, namely Random Forest, K-Nearest Neighbors, Support Vector Machine, Multi-Layer Perceptron, for enhanced anomaly detection. FusionNet's architecture leverages diversity these algorithms to achieve high accuracy precision. We evaluate performance on two distinct datasets, Dataset 1 2, compare it with traditional models, SVM, KNN, RF. The results demonstrate FusionNet consistently outperforms models across both datasets in terms accuracy, precision, recall, F1 score. On 1, achieves 98.5% attains 99.5%. remarkable ability detect anomalies exceptional underscores its potential real-world applications. This study highlights significance as robust provides insights into superior over models. emphasize promising prospects other domains where accurate crucial.

Language: Английский

---

Advanced artificial intelligence with federated learning framework for privacy-preserving cyberthreat detection in IoT-assisted sustainable smart cities

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

Published: Feb. 6, 2025

Language: Английский

---

Analysis of convolutional-based variational autoencoders for privacy protection in realtime video surveillance

Expert Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 126817 - 126817

Published: Feb. 1, 2025

Language: Английский

---

BFLIDS: Blockchain-Driven Federated Learning for Intrusion Detection in IoMT Networks

Sensors, Journal Year: 2024, Volume and Issue: 24(14), P. 4591 - 4591

Published: July 15, 2024

The Internet of Medical Things (IoMT) has significantly advanced healthcare, but it also brought about critical security challenges. Traditional solutions struggle to keep pace with the dynamic and interconnected nature IoMT systems. Machine learning (ML)-based Intrusion Detection Systems (IDS) have been increasingly adopted counter cyberattacks, centralized ML approaches pose privacy risks due single points failure (SPoFs). Federated Learning (FL) emerges as a promising solution, enabling model updates directly on end devices without sharing private data central server. This study introduces BFLIDS, Blockchain-empowered Learning-based IDS designed enhance intrusion detection in networks. Our approach leverages blockchain secure transaction records, FL maintain by training models locally, IPFS for decentralized storage, MongoDB efficient management. Ethereum smart contracts (SCs) oversee all interactions transactions within system. We modified FedAvg algorithm Kullback-Leibler divergence estimation adaptive weight calculation boost accuracy robustness against adversarial attacks. For classification, we implemented an Adaptive Max Pooling-based Convolutional Neural Network (CNN) Bidirectional Long Short-Term Memory (BiLSTM) attention residual connections Edge-IIoTSet TON-IoT datasets. achieved accuracies 97.43% (for CNNs Edge-IIoTSet), 96.02% BiLSTM 98.21% TON-IoT), 97.42% TON-IoT) scenarios, which are competitive methods. proposed BFLIDS effectively detects intrusions, enhancing

Language: Английский

---

A data-driven multi-perspective approach to cybersecurity knowledge discovery through topic modelling

Alexandria Engineering Journal, Journal Year: 2024, Volume and Issue: 107, P. 374 - 389

Published: July 24, 2024

Language: Английский

---

Blockchain-Integrated Security for Real-Time Patient Monitoring in the Internet of Medical Things Using Federated Learning

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 117826 - 117850

Published: Jan. 1, 2023

The Internet of Medical Things (IoMT) heralds a transformative era in healthcare, with the potential to revolutionize patient care, healthcare services, and medical research. As all technological progressions, IoMT introduces suite complex challenges, predominantly centered on security. In particular, ensuring integrity, confidentiality, availability health data real-time communication stands paramount, given sensitivity information ramifications breaches or misuse. light these existing security frameworks, while commendable, exhibit limitations. Specifically, they often grapple comprehensive anomaly detection, effective resistance replay attacks, robust protection against threats like man-in-the-middle eavesdropping, tampering, identity spoofing. proposed framework integrates state-of-the-art encryption techniques, cutting-edge pattern recognition modules, adaptive learning mechanisms. These components collaboratively ensure integrity during transmission, provide conventional novel attack vectors, adapt evolving through continuous learning. Moreover, incorporates sophisticated checksum techniques advanced behavioral analysis, further enhancing its protective capabilities. Our system demonstrated significant improvements detection metrics, consistently outperforming benchmark solutions MRMS BACKM-EHA.

Language: Английский

---

Big-IDS: a decentralized multi agent reinforcement learning approach for distributed intrusion detection in big data networks

Cluster Computing, Journal Year: 2024, Volume and Issue: 27(5), P. 6823 - 6841

Published: March 8, 2024

Language: Английский

---

WFE-Tab: Overcoming limitations of TabPFN in IIoT-MEC environments with a weighted fusion ensemble-TabPFN model for improved IDS performance

Future Generation Computer Systems, Journal Year: 2025, Volume and Issue: unknown, P. 107707 - 107707

Published: Jan. 1, 2025

Language: Английский

---