Attacks Detection and Mitigation of IoT Using Machine Learning Model

Advances in computational intelligence and robotics book series, Год журнала: 2025, Номер unknown, С. 115 - 132

Опубликована: Фев. 28, 2025

The RPL protocol has been increasingly used in low-loss networks and Internet of Things recent years. is exposed to many attacks that affect the network performance disrupt connection. In this study, a machine learning-based intrusion detection system (IDS) was developed distinguish between benign malicious attacks. Ensemble learning applied by stacking three models: Random Forest, Gradient Boosting, Linear SVC with hyperparameter optimization using search CV. proposed approach version number, Decreased rank, Hello Flooding database. results obtained indicate provides reliable effective solution protect from attacks, it also sought achieve balance accuracy computational efficiency. model achieved an 98% for HF database, 97% VN 95% DR

Язык: Английский

---

Analysis for Detection and Mitigation of Version Number Modification Attack in the Internet of Things

Advances in computational intelligence and robotics book series, Год журнала: 2025, Номер unknown, С. 133 - 150

Опубликована: Фев. 28, 2025

The connectivity of devices and the ability to collect massive amounts data, Internet Things (IoT) has radically changed networking. However, security privacy issues still exist, especially in networks with limited resources. In IoT applications, RPL routing protocol is frequently used for routing; version number attacks represent a issue. Considering sensor node mobility evolved into IoET (internet Everything), which encompasses all electronic devices, including body networks, VANETs, smart grid stations. This paper studies effects on networks. To detect rogue nodes, save overhead, improve lossy low-power modification approach proposed as well demonstrate how create test these using COOJA Contiki-NG.

Язык: Английский

---

Enhancing IoT Security RPL Attack Detection Using Sine Cosine Algorithm With XGBoost

Advances in information security, privacy, and ethics book series, Год журнала: 2025, Номер unknown, С. 1 - 28

Опубликована: Янв. 17, 2025

The most widely used routing protocol for Internet of Things (IoT) networks with limited resources is the (RPL). Network security a significant issue because exponential growth devices and their growing ubiquity in safety-critical settings like healthcare industry. One possible remedy threat identification these using intrusion detection systems (IDS) based on machine learning. To achieve this, learning approach that uses Random Forest (RF), k-nearest Neighbor (KNN), Decision Tree (DT), XGBoost, Support Vector Machine (SVM) models presented. proposed learning-based conducts mitigate feature classification Sine Cosine Algorithm (SCA) XGBoost to select smallest number relevant features, leading best solution highest accuracy. model achieves high accuracy 97% Decrease Rank Version Number. Also, 99% Hello Flooding.

Язык: Английский

---

Security Concerns with IoT Routing: A Review of Attacks, Countermeasures, and Future Prospects

Advances in Internet of Things, Год журнала: 2024, Номер 14(04), С. 67 - 98

Опубликована: Янв. 1, 2024

Язык: Английский

---