FL-DSFA: Securing RPL-Based IoT Networks against Selective Forwarding Attacks Using Federated Learning DOI Creative Commons

Rabia Khan,

Noshina Tariq, Muhammad Imran Ashraf

и другие.

Sensors, Год журнала: 2024, Номер 24(17), С. 5834 - 5834

Опубликована: Сен. 8, 2024

The Internet of Things (IoT) is a significant technological advancement that allows for seamless device integration and data flow. development the IoT has led to emergence several solutions in various sectors. However, rapid popularization also its challenges, one most serious challenges security IoT. Security major concern, particularly routing attacks core network, which may cause severe damage due information loss. Routing Protocol Low-Power Lossy Networks (RPL), protocol used devices, faced with selective forwarding attacks. In this paper, we present federated learning-based detection technique detecting attacks, termed FL-DSFA. A lightweight model involving Attack Dataset (IRAD), comprises Hello Flood (HF), Decreased Rank (DR), Version Number (VN), increase efficiency. on threaten system since they mainly focus essential elements RPL. components include control messages, topologies, repair procedures, resources within sensor networks. Binary classification approaches have been assess training efficiency proposed model. step includes implementation machine learning algorithms, including logistic regression (LR), K-nearest neighbors (KNN), support vector (SVM), naive Bayes (NB). comparative analysis illustrates study, SVM KNN classifiers, exhibits highest accuracy during achieves efficient runtime performance. demonstrates exceptional performance, achieving prediction precision 97.50%, an 95%, recall rate 98.33%, F1 score 97.01%. It outperforms current leading research field, results, scalability, enhanced privacy.

Язык: Английский

FL-DSFA: Securing RPL-Based IoT Networks against Selective Forwarding Attacks Using Federated Learning DOI Creative Commons

Rabia Khan,

Noshina Tariq, Muhammad Imran Ashraf

и другие.

Sensors, Год журнала: 2024, Номер 24(17), С. 5834 - 5834

Опубликована: Сен. 8, 2024

The Internet of Things (IoT) is a significant technological advancement that allows for seamless device integration and data flow. development the IoT has led to emergence several solutions in various sectors. However, rapid popularization also its challenges, one most serious challenges security IoT. Security major concern, particularly routing attacks core network, which may cause severe damage due information loss. Routing Protocol Low-Power Lossy Networks (RPL), protocol used devices, faced with selective forwarding attacks. In this paper, we present federated learning-based detection technique detecting attacks, termed FL-DSFA. A lightweight model involving Attack Dataset (IRAD), comprises Hello Flood (HF), Decreased Rank (DR), Version Number (VN), increase efficiency. on threaten system since they mainly focus essential elements RPL. components include control messages, topologies, repair procedures, resources within sensor networks. Binary classification approaches have been assess training efficiency proposed model. step includes implementation machine learning algorithms, including logistic regression (LR), K-nearest neighbors (KNN), support vector (SVM), naive Bayes (NB). comparative analysis illustrates study, SVM KNN classifiers, exhibits highest accuracy during achieves efficient runtime performance. demonstrates exceptional performance, achieving prediction precision 97.50%, an 95%, recall rate 98.33%, F1 score 97.01%. It outperforms current leading research field, results, scalability, enhanced privacy.

Язык: Английский

Процитировано

0