Distributed Denial of Services (DDoS) attack detection in SDN using Optimizer-equipped CNN-MLP

PLoS ONE, Год журнала: 2025, Номер 20(1), С. e0312425 - e0312425

Опубликована: Янв. 27, 2025

Software-Defined Networks (SDN) provides more control and network operation over a infrastructure as an emerging revolutionary paradigm in networking. Operating the many applications preserving services functions, SDN controller is regarded operating system of SDN-based architecture. The has several security problems because its intricate design, even with all amazing features. Denial-of-service (DoS) attacks continuously impact users Internet service providers (ISPs). Because centralized distributed denial (DDoS) on are frequent may have widespread effect network, particularly at layer. We propose to implement both MLP (Multilayer Perceptron) CNN (Convolutional Neural Networks) based conventional methods detect Denial Services attack. These models got complex optimizer installed them decrease false positive or DDoS case detection efficiency. use SHAP feature selection technique improve procedure. By assisting identification which features most essential spot incidents, approach aids process enhancing precision flammability. Fine-tuning hyperparameters help Bayesian optimization obtain best model performance another important thing that we do our model. Two datasets, InSDN CICDDoS-2019, utilized assess effectiveness proposed method, 99.95% for true (TP) CICDDoS-2019 dataset 99.98% dataset, results show highly accurate.

Язык: Английский

---

Detection of DDoS attacks in SDN-based VANET using optimized TabNet

Computer Standards & Interfaces, Год журнала: 2024, Номер 90, С. 103845 - 103845

Опубликована: Фев. 18, 2024

Язык: Английский

---

Detection of DDoS Attack using Machine Learning Algorithms

International Journal for Research in Applied Science and Engineering Technology, Год журнала: 2024, Номер 12(3), С. 1511 - 1523

Опубликована: Март 20, 2024

Abstract: The utilization of the internet has greatly increased in recent decades, leading to a vulnerability networking and cybersecurity. One most common resulting attacks is Distributed Denial Service (DDoS), where overwhelming amounts data are sent legitimate websites or servers, causing delays denying access users. Single source known as denial service (DoS), while from multiple sources, such botnet, considered distributed (DDoS). In our project, we employed three machine learning algorithms identify DDoS attacks, determined successful algorithm based on accuracy metric. We trained tested using standardized dataset, dataset_sdn, obtained experimental results. Out all used, XGBoost proved be effective with an 99.9%. During preprocessing, any missing was replaced column's mean value

Язык: Английский

---

Improvement of Distributed Denial of Service Attack Detection through Machine Learning and Data Processing

Mathematics, Год журнала: 2024, Номер 12(9), С. 1294 - 1294

Опубликована: Апрель 25, 2024

The early and accurate detection of Distributed Denial Service (DDoS) attacks is a fundamental area research to safeguard the integrity functionality organizations’ digital ecosystems. Despite growing importance neural networks in recent years, use classical techniques remains relevant due their interpretability, speed, resource efficiency, satisfactory performance. This article presents results comparative analysis six machine learning techniques, namely, Random Forest (RF), Decision Tree (DT), AdaBoost (ADA), Extreme Gradient Boosting (XGB), Multilayer Perceptron (MLP), Dense Neural Network (DNN), for classifying DDoS attacks. CICDDoS2019 dataset was used, which underwent data preprocessing remove outliers, 22 features were selected using Pearson correlation coefficient. RF classifier achieved best accuracy rate (99.97%), outperforming other classifiers even previously published network-based techniques. These findings underscore feasibility effectiveness algorithms field attack detection, reaffirming relevance as valuable tool advanced cyber defense.

Язык: Английский

---

A new DDoS attack detection model based on improved stacked autoencoder and gated recurrent unit for software defined network

The Computer Journal, Год журнала: 2025, Номер unknown

Опубликована: Март 11, 2025

Abstract With the widespread adoption of Software Defined Networking (SDN), detecting Distributed Denial Service (DDoS) attacks has become an urgent challenge in SDN maintenance and Security. Given diversity DDoS attack types, we face significant challenges. This paper proposes a model called ARSAE-QGRU, which is based on integrating attention mechanisms residual connections within stacked autoencoder for detection. By introducing into (SAE), effectively conveys more valuable information facilitates gradient propagation, allowing it to learn low-dimensional representations better. It also combines learned with traffic features generate data training. Furthermore, incorporating Gated Recurrent Unit aids in-depth understanding temporal characteristics data, resulting improved detection accuracy. demonstrates outstanding performance CICDDoS2019 CICIDS2017 datasets, achieving accuracy rates 97.2% 97.9%, respectively. Moreover, when applied datasets environments, reaches even higher rate 99.8%. research provides reliable solution high-dimensional processing SDN, addressing challenges these domains.

Язык: Английский

---

Enhancing Threat Detection in Financial Cyber Security Through Auto Encoder-MLP Hybrid Models

International Journal of Advanced Computer Science and Applications, Год журнала: 2024, Номер 15(4)

Опубликована: Янв. 1, 2024

Cyber-attacks have the potential to cause power outages, malfunctions with military equipment, and breaches of sensitive data. Owing substantial financial value information it contains, banking sector is especially vulnerable. The number digital footprints that banks increases, increasing attack surface available hackers. This paper presents a unique approach improve cyber security threat detection by integrating Auto Encoder-Multilayer Perceptron (AE-MLP) hybrid models. These models use MLP neural networks' discriminative capabilities for tasks, while also utilizing auto encoders' strengths in collecting complex patterns abnormalities NSL-KDD dataset, which varied includes transaction records, user activity patterns, network traffic, was thoroughly analysed. results show AE-MLP perform well spotting possible risks including fraud, data breaches, unauthorized access attempts. encoders accuracy methods efficiently compressing rebuilding complicated representations. makes easier extract latent characteristics are essential differentiating between normal abnormal activity. implemented Python software. recommended Hybrid AE+MLP shows better 99%, 13.16% more sophisticated, when compared traditional approach. suggested improves systems' capacity prediction providing scalability efficiency handling massive amounts real-time settings.

Язык: Английский

---

Detecting DDoS based on attention mechanism for Software-Defined Networks

Journal of Network and Computer Applications, Год журнала: 2024, Номер 230, С. 103928 - 103928

Опубликована: Июнь 25, 2024

Язык: Английский

---

Reflective Distributed Denial of Service Detection: A Novel Model Utilizing Binary Particle Swarm Optimization—Simulated Annealing for Feature Selection and Gray Wolf Optimization-Optimized LightGBM Algorithm

Sensors, Год журнала: 2024, Номер 24(19), С. 6179 - 6179

Опубликована: Сен. 24, 2024

The fast growth of the Internet has made network security problems more noticeable, so intrusion detection systems (IDSs) have become a crucial tool for maintaining security. IDSs guarantee normal operation by tracking traffic and spotting possible assaults, thereby safeguarding data However, traditional methods encounter several issues such as low efficiency prolonged time when dealing with massive high-dimensional data. Therefore, feature selection (FS) is particularly important in IDSs. By selecting most representative features, it can not only improve accuracy but also significantly reduce computational complexity attack time. This work proposes new FS approach, BPSO-SA, that based on Binary Particle Swarm Optimization (BPSO) Simulated Annealing (SA) algorithms. It combines these Gray Wolf (GWO) algorithm to optimize LightGBM model, building type reflective Distributed Denial Service (DDoS) model. BPSO-SA enhances global search capability (PSO) using SA mechanism effectively screens out optimal subset; GWO optimizes hyperparameters simulating group hunting behavior gray wolves enhance performance While showing great resilience generalizing power, experimental results show proposed DDoS model surpasses conventional terms accuracy, precision, recall, F1-score, prediction

Язык: Английский

---

Reinforcing Network Security: Network Attack Detection Using Random Grove Blend in Weighted MLP Layers

Mathematics, Год журнала: 2024, Номер 12(11), С. 1720 - 1720

Опубликована: Май 31, 2024

In the modern world, evolution of internet supports automation several tasks, such as communication, education, sports, etc. Conversely, it is prone to types attacks that disturb data transfer in network. Efficient attack detection needed avoid consequences an attack. Traditionally, manual limited by human error, less efficiency, and a time-consuming mechanism. To address problem, large number existing methods focus on techniques for better efficacy detection. However, improvement significant factors accuracy, handling larger data, over-fitting versus fitting, tackle this issue, proposed system utilized Random Grove Blend Weighted MLP (Multi-Layer Perceptron) Layers classify network attacks. The used its advantages solving complex non-linear problems, datasets, high accuracy. computation requirements great deal labeled training data. resolve random info grove blend weight weave layer are incorporated into attain this, UNSW–NB15 dataset, which comprises nine attack, detect Moreover, Scapy tool (2.4.3) generate real-time dataset classifying efficiency presented mechanism calculated with performance metrics. Furthermore, internal external comparisons processed respective research reveal system’s efficiency. model utilizing attained accuracy 98%. Correspondingly, intended contribute associated enhancing security.

Язык: Английский

---

A comprehensive plane-wise review of DDoS attacks in SDN: Leveraging detection and mitigation through machine learning and deep learning

Journal of Network and Computer Applications, Год журнала: 2024, Номер 235, С. 104081 - 104081

Опубликована: Дек. 9, 2024

Язык: Английский

---