Enhancing IoT security: A Creative Swagger Optimization algorithm for DDoS defence DOI
Rahul Papalkar, A. S. Alvi

Network Computation in Neural Systems, Год журнала: 2024, Номер unknown, С. 1 - 39

Опубликована: Дек. 30, 2024

In the Internet of Things (IoT), security information between network transmissions is very important since system stores data in storage and performed by exchange about things. DDoS an IoT attack that targets availability servers flooding communication channel with impersonated requests coming from distributed devices. To overcome above-mentioned issue, this research proposed a Creative Swagger (CS) Optimized Deep Convolutional Neural Network (DeepCNN) detects mitigates attacks. The CS algorithm designed fusing distinctive behaviour innovative concepts civilized creature, which used to effectively tune parameters CNN improve detection accuracy For initial verification, blacklist table verification includes checking IP address other pertinent attributes. CS-optimized model obtains high effectiveness attaining 97.07%, sensitivity 97.23%, specificity 96.91% at 80% training for utilizing UNSW-NB15 Dataset. Moreover, method provides best solution detecting attacks platforms higher robustness.

Язык: Английский

A hybrid approach for efficient feature selection in anomaly intrusion detection for IoT networks DOI Creative Commons
Aya G. Ayad, Nehal A. Sakr, Noha A. Hikal

и другие.

The Journal of Supercomputing, Год журнала: 2024, Номер 80(19), С. 26942 - 26984

Опубликована: Авг. 29, 2024

Abstract The exponential growth of Internet Things (IoT) devices underscores the need for robust security measures against cyber-attacks. Extensive research in IoT community has centered on effective traffic detection models, with a particular focus anomaly intrusion systems (AIDS). This paper specifically addresses preprocessing stage datasets and feature selection approaches to reduce complexity data. goal is develop an efficient AIDS that strikes balance between high accuracy low time. To achieve this goal, we propose hybrid approach combines filter wrapper methods. integrated into two-level system. At level 1, our classifies network packets normal or attack, 2 further classifying attack determine its specific category. One critical aspect consider imbalance these datasets, which addressed using Synthetic Minority Over-sampling Technique (SMOTE). evaluate how selected features affect performance machine learning model across different algorithms, namely Decision Tree, Random Forest, Gaussian Naive Bayes, k-Nearest Neighbor, employ benchmark datasets: BoT-IoT, TON-IoT, CIC-DDoS2019. Evaluation metrics encompass accuracy, precision, recall, F1-score. Results indicate decision tree achieves ranging 99.82 100%, short times 0.02 0.15 s, outperforming existing architectures networks establishing superiority achieving both times.

Язык: Английский

Процитировано

7
Identificación de ataques de denegación de servicio distribuido (DDoS) mediante la integración de algoritmos de aprendizaje automático y arquitecturas de redes neuronales artificiales. DOI Creative Commons
Víctor Alfonso Guzmán Brand, Laura Esperanza Gélvez García

Revista de Ingeniería Matemáticas y Ciencias de la Información, Год журнала: 2025, Номер 12(23)

Опубликована: Янв. 31, 2025

Objective: To identify distributed denial of service (DDoS) attacks by integrating machine learning algorithms and artificial neural network architectures. Methodology: structure the data analysis, Knowledge Discovery Data (KDD) technique is used. This approach allows examining large volumes information various types, with objective identifying patterns, correlations producing valuable information. As for set, CIC-DDoS2019 dataset developed Canadian Cybersecurity Institute Results: When training evaluating different algorithms, it was observed that models based on decision trees, such as Random Forest XGBoost, stood out achieving best results in terms accuracy efficiency. On other hand, analysis performance networks, Closed Stream Units (GRU) obtaining precision. suggests GRUs achieve an optimal balance between predictive ability minimization false positives negatives. Discussion: In comparison traditional networks DDoS attack detection, XGBoost offer similar or superior also exhibit significantly shorter execution times. GRU RNN high accuracy, but a computational cost. Conclusions: demonstrated (F1-score: 0.9992) speed (11.47s), positioning itself most viable alternative real-time implementations. field Gated (GCU) obtained (accuracy: 0.9992; F1-score: 0.9992), given to process temporal dependencies reduce positives.

Язык: Английский

Процитировано

0
Heuristically enhanced multi-head attention based recurrent neural network for denial of wallet attacks detection on serverless computing environment DOI Creative Commons
Sarah A. Alzakari, Mohammad Alamgeer, Abdullah Mujawib Alashjaee

и другие.

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

Опубликована: Апрель 19, 2025

Denial of Wallet (DoW) attacks are a cyber threat designed to utilize and deplete an organization's financial resources by generating excessive prices or charges in their cloud computing (CC) serverless platforms. These threats primarily appropriate manners because features such as auto-scaling, pay-as-you-go, restricted control, cost growth. Serverless computing, frequently recognized Function-as-a-Service (FaaS), is CC method that permits designers construct run uses without the requirement accomplish typical server structure. Detecting DoW involves monitoring analyzing system-level resource consumption specific bare-metal mechanisms. Efficient precise detection internal remains crucial challenge. Timely recognition significant preventing potential damage, exploit model environments, impacting structure operational integrity services. In this study, Multi-Head Attention-based Recurrent Neural Network for Attacks Detection (MHARNN-DoWAD) technique developed. The MHARNN-DoWAD enables on environments. At first, presented performs data preprocessing using min-max normalization convert input into constant format. Next, wolf pack predation (WPP) employed feature selection. classification attacks, multi-head attention-based bi-directional gated recurrent unit (MHA-BiGRU) utilized. Eventually, improved secretary bird optimizer algorithm (ISBOA)-based hyperparameter choice process accomplished optimize results MHA-BiGRU model. A comprehensive set simulations was conducted demonstrate promising method. experimental validation portrayed superior accuracy value 98.30% over existing models.

Язык: Английский

Процитировано

0
Detecting DDoS Attacks Through Decision Tree Analysis: An EDA Approach with the CIC DDoS 2019 Dataset DOI
Ahmad Turmudi Zy,

Amali,

Anggi Muhammad Rifa’i

и другие.

Опубликована: Авг. 29, 2024

Язык: Английский

Процитировано

1
Enhancing IoT security: A Creative Swagger Optimization algorithm for DDoS defence DOI
Rahul Papalkar, A. S. Alvi

Network Computation in Neural Systems, Год журнала: 2024, Номер unknown, С. 1 - 39

Опубликована: Дек. 30, 2024

In the Internet of Things (IoT), security information between network transmissions is very important since system stores data in storage and performed by exchange about things. DDoS an IoT attack that targets availability servers flooding communication channel with impersonated requests coming from distributed devices. To overcome above-mentioned issue, this research proposed a Creative Swagger (CS) Optimized Deep Convolutional Neural Network (DeepCNN) detects mitigates attacks. The CS algorithm designed fusing distinctive behaviour innovative concepts civilized creature, which used to effectively tune parameters CNN improve detection accuracy For initial verification, blacklist table verification includes checking IP address other pertinent attributes. CS-optimized model obtains high effectiveness attaining 97.07%, sensitivity 97.23%, specificity 96.91% at 80% training for utilizing UNSW-NB15 Dataset. Moreover, method provides best solution detecting attacks platforms higher robustness.

Язык: Английский

Процитировано

0