TY - GEN
T1 - NIDS
T2 - 2022 International Conference on Virtual Reality, Human-Computer Interaction and Artificial Intelligence, VRHCIAI 2022
AU - Mohi-Ud-Din, Ghulam
AU - Zheng, Jiangbin
AU - Liu, Zhiqiang
AU - Asim, Muhammad
AU - Chen, Jiajun
AU - Liu, Jinjing
AU - Lin, Zhijun
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Network traffic data has developed into an appealing target for attacks as new network communication services have drawn more attention. Due to the enormous number of data these systems create, conventional intrusion detection systems (IDS) cannot detect intruder behaviors in large-scale network systems. To detect malicious assaults early on, an effective IDS must be able to scan massive volumes of network traffic data quickly. In order to identify various types of intrusion in network systems, this study introduces a unique distributed network IDS (NIDS). The suggested NIDS is built on a distributed random forest that can analyze huge amounts of data quickly. The two steps of the suggested method are the traffic gathering module and the attack detection module. In VANETs, the random forest method was employed for real-time DDoS attack detection. A variety of performance criteria, including accuracy, precision, recall, and F1 score, were tested on the system to confirm the performance of the suggested framework. The proposed method achieves improved classification accuracy, according to the results. The suggested approach is suitable for complicated systems with high speed and low false alarm rates that require real-time intrusion detection.
AB - Network traffic data has developed into an appealing target for attacks as new network communication services have drawn more attention. Due to the enormous number of data these systems create, conventional intrusion detection systems (IDS) cannot detect intruder behaviors in large-scale network systems. To detect malicious assaults early on, an effective IDS must be able to scan massive volumes of network traffic data quickly. In order to identify various types of intrusion in network systems, this study introduces a unique distributed network IDS (NIDS). The suggested NIDS is built on a distributed random forest that can analyze huge amounts of data quickly. The two steps of the suggested method are the traffic gathering module and the attack detection module. In VANETs, the random forest method was employed for real-time DDoS attack detection. A variety of performance criteria, including accuracy, precision, recall, and F1 score, were tested on the system to confirm the performance of the suggested framework. The proposed method achieves improved classification accuracy, according to the results. The suggested approach is suitable for complicated systems with high speed and low false alarm rates that require real-time intrusion detection.
KW - DDoS attacks (keywords)
KW - Network Intrusion Detection Systems
KW - Random Forest
KW - Vehicular Ad Hoc Networks
UR - http://www.scopus.com/inward/record.url?scp=85158831540&partnerID=8YFLogxK
U2 - 10.1109/VRHCIAI57205.2022.00051
DO - 10.1109/VRHCIAI57205.2022.00051
M3 - 会议稿件
AN - SCOPUS:85158831540
T3 - Proceedings - 2022 International Conference on Virtual Reality, Human-Computer Interaction and Artificial Intelligence, VRHCIAI 2022
SP - 255
EP - 260
BT - Proceedings - 2022 International Conference on Virtual Reality, Human-Computer Interaction and Artificial Intelligence, VRHCIAI 2022
A2 - Ma, Nan
A2 - Liao, Miao
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 28 October 2022 through 30 October 2022
ER -