SDN-based detection and mitigation of DDoS attacks on smart homes

Computer Communications, Год журнала: 2024, Номер 221, С. 29 - 41

Опубликована: Апрель 3, 2024

The adoption of the Internet Things (IoT) has proliferated across various domains, where everyday objects like refrigerators and washing machines are now equipped with sensors connected to internet. Undeniably, security such devices, which were not primarily designed for internet connectivity, is utmost importance but been largely neglected. In this paper, we propose a framework real-time DDoS attack detection mitigation in SDN-enabled smart home networks. We capture network traffic during regular operations attacks. This captured used train several machine learning (ML) models, including Support Vector Machine (SVM), Logistic Regression, Decision Trees, K-Nearest Neighbors (KNN) algorithms. These trained models executed as SDN controller applications subsequently employed detection. While utilize ML techniques protect IoT use SNORT, signature-based technique, secure itself. Real-world experiments demonstrate that without goes offline shortly after an attack, resulting 100% packet loss. Furthermore, show algorithms can efficiently classify into benign traffic, Tree algorithm outperforming others accuracy 99%.

Язык: Английский

---

A Genetic Algorithm- and t-Test-Based System for DDoS Attack Detection in IoT Networks

IEEE Access, Год журнала: 2024, Номер 12, С. 25623 - 25641

Опубликована: Янв. 1, 2024

Internet and cloud-based technologies have facilitated the implementation of large-scale Things (IoT) networks. However, these networks are susceptible to emerging attacks. This paper proposes a novel lightweight system for detecting both high- low-volume Distributed Denial Service (DDoS) attacks in IoT networks, namely Genetic Algorithm (GA) t-Test DDoS Attack Detection (GADAD). The GADAD employs edge-based has three phases. In first phase, it creates preprocesses an HL-IoT (High- Low-volume networks) dataset, which includes second phase introduces method, called GAStats, optimal feature selection using GA statistical parameters (Stats.). third trains tree-based Machine Learning (ML) models: Random Forest (RF), Extra-Tree (ET), Adaptive Boosting (AdaBoost), along with other ML models, self-generated dataset publicly available ToN-IoT dataset. evaluation assessment key performance metrics such as accuracy, precision, recall, F1-score, Receiver Operating Characteristic Curve (ROC), computation time, scalability analysis overall performance. experimental results illustrate efficacy method optimizing system's efficiency reduction time compared existing state-of-the-art techniques.

Язык: Английский

---

An empirical study of pattern leakage impact during data preprocessing on machine learning-based intrusion detection models reliability

Expert Systems with Applications, Год журнала: 2023, Номер 230, С. 120715 - 120715

Опубликована: Июнь 8, 2023

Язык: Английский

---

BukaGini: A Stability-Aware Gini Index Feature Selection Algorithm for Robust Model Performance

IEEE Access, Год журнала: 2023, Номер 11, С. 59386 - 59396

Опубликована: Янв. 1, 2023

Feature interaction is a vital aspect of Machine Learning (ML) algorithms, and gaining deep understanding these interactions can significantly enhance model performance. This paper introduces the BukaGini algorithm, an innovative robust approach for feature analysis that capitalizes on Gini impurity index. By exploiting unique properties index, our proposed algorithm effectively captures both linear nonlinear interactions, providing richer more comprehensive representation underlying data. We thoroughly evaluate against traditional index-based methods various real-world datasets. These datasets include High School Students' Performance (HSSP) dataset, which examines factors affecting student performance; Cancer Data, focuses identifying cancer types based gene expression; Spambase, targets spam email classification; UNSW-NB15 addresses network intrusion detection. Our experimental results demonstrate consistently outperforms in terms accuracy. Across tested datasets, achieves improvements ranging from 0.32% to 2.50%, underscoring its effectiveness handling diverse data problem domains. performance gain highlights potential as valuable tool ML applications.

Язык: Английский

---

Overcoming the Challenges of Data Lack, Leakage, and Dimensionality in Intrusion Detection Systems: A Comprehensive Review

Journal of Communication and Information Systems, Год журнала: 2024, Номер 39(2024), С. 22 - 34

Опубликована: Янв. 1, 2024

The Internet of Things (IoT) and cloud computing are rapidly gaining momentum as decentralized internet-based technologies have led to an increase in information nearly every technical commercial industry. However, ensuring the security IoT systems is a pressing issue due complexities involved connected shared environments. Networks guarded by Intrusion Detection Systems (IDS) against various cyber threats such malware, viruses, unauthorized access. IDS recently adopted Machine Learning (ML) Deep (DL) techniques identify classify risks. effective utilization these depends on availability, quality, characteristics data used train models. Moreover, lack, leak, dimensionality (DLLD) common problems science ML. This paper surveys existing research suggests solutions for overcoming DLLD-related issues improve model.

Язык: Английский

---

SMRD: A Novel Cyber Warfare Modeling Framework for Social Engineering, Malware, Ransomware, and Distributed Denial-of-Service Based on a System of Nonlinear Differential Equations

Journal of Applied Artificial Intelligence, Год журнала: 2024, Номер 5(1)

Опубликована: Март 20, 2024

Cyber warfare has emerged as a critical aspect of modern conflict, state and non-state actors increasingly leverage cyber capabilities to achieve strategic objectives. The rapidly evolving threat landscape demands robust adaptive approaches protect against advanced cyberattacks mitigate their impact on national security. Traditional defense strategies often struggle keep pace with the changing landscape, resulting in need for more cyberattacks. This paper presents novel modeling framework, Social Engineering, Malware, Ransomware, Distributed Denial-of-Service (SMRD), capturing interactions interdependencies between these core components. SMRD framework offers insights enhancing defense, prediction, proactive measures. A mathematical model consisting system nonlinear differential equations is proposed quantify relationships dynamics

Язык: Английский

---

Application of BukaGini algorithm for enhanced feature interaction analysis in intrusion detection systems

PeerJ Computer Science, Год журнала: 2024, Номер 10, С. e2043 - e2043

Опубликована: Апрель 30, 2024

This article presents an evaluation of BukaGini, a stability-aware Gini index feature selection algorithm designed to enhance model performance in machine learning applications. Specifically, the study focuses on assessing BukaGini’s effectiveness within domain intrusion detection systems (IDS). Recognizing need for improved interaction analysis methodologies IDS, this research aims investigate BukaGini context. is evaluated across four diverse datasets commonly used IDS research: NSLKDD (22,544 samples), WUSTL EHMS (16,318 WSN-DS (374,661 and UNSWNB15 (175,341 amounting total 588,864 data samples. The encompasses key metrics such as stability score, accuracy, F1-score, recall, precision, ROC AUC. Results indicate significant advancements performance, with achieving remarkable accuracy rates up 99% scores consistently surpassing all datasets. Additionally, demonstrates average reduction dimensionality 25%, selecting 10 features each dataset using index. Through rigorous comparative existing methodologies, emerges promising solution cybersecurity applications, particularly context IDS. These findings highlight potential contribute robust propel capabilities new heights real-world scenarios.

Язык: Английский

---

HAE-HRL: A network intrusion detection system utilizing a novel autoencoder and a hybrid enhanced LSTM-CNN-based residual network

Computers & Security, Год журнала: 2025, Номер unknown, С. 104328 - 104328

Опубликована: Янв. 1, 2025

Язык: Английский

---

A new a flow-based approach for enhancing botnet detection using convolutional neural network and long short-term memory

Knowledge and Information Systems, Год журнала: 2025, Номер unknown

Опубликована: Апрель 16, 2025

Язык: Английский

---

Distributed denial-of-service (DDOS) attack detection using supervised machine learning algorithms

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Апрель 16, 2025

Язык: Английский

---