TY - GEN
T1 - TLP-IDS
T2 - 39th International Symposium on Reliable Distributed Systems, SRDS 2020
AU - Liu, Xiaoxia
AU - He, Daojing
AU - Gao, Yun
AU - Zhu, Sencun
AU - Chan, Sammy
N1 - Funding Information:
This research is supported by the National Natural Science Foundation of China (Grants: U1936120, U1636216), the National Key R&D Program of China (2017YFB0802805 and 2017YFB0801701), Joint Fund of Ministry of Education of China for Equipment Preresearch (No. 6141A020333), the Fundamental Research Funds for the Central Universities, and the Basic Research Program of State Grid Shanghai Municipal Electric Power Company (52094019007F). Daojing He is the corresponding author of this article.
Publisher Copyright:
© 2020 IEEE.
PY - 2020/9
Y1 - 2020/9
N2 - With the increasing applications of integrated electronic systems (IESs), especially in security critical application scenarios like satellites and aircraft, new vulnerabilities and attacks have emerged recently. To detect the attacks, we propose TLP-IDS, a real-time intrusion detection system (IDS). TLP-IDS includes two layers of detection modules, one based on time and sequence logic and the other based on historical data. For the modules in the first layer, periodic and aperiodic messages are distinguished based on variations of message intervals, and we learnd from the idea of Markov decision process (MDP) in reinforcement learning (RL) to automatically learn the logical relationship between sequences. In the second layer, an online sequence extreme learning machine (OS-ELM) method is deployed to fit the data and further combined with the Weibull distribution function for prediction and detection. To evaluate our system, we implement several attack scenarios on a test bed, and measure the detection performance. Experimental results show that our system can quickly and effectively detect various attacks.
AB - With the increasing applications of integrated electronic systems (IESs), especially in security critical application scenarios like satellites and aircraft, new vulnerabilities and attacks have emerged recently. To detect the attacks, we propose TLP-IDS, a real-time intrusion detection system (IDS). TLP-IDS includes two layers of detection modules, one based on time and sequence logic and the other based on historical data. For the modules in the first layer, periodic and aperiodic messages are distinguished based on variations of message intervals, and we learnd from the idea of Markov decision process (MDP) in reinforcement learning (RL) to automatically learn the logical relationship between sequences. In the second layer, an online sequence extreme learning machine (OS-ELM) method is deployed to fit the data and further combined with the Weibull distribution function for prediction and detection. To evaluate our system, we implement several attack scenarios on a test bed, and measure the detection performance. Experimental results show that our system can quickly and effectively detect various attacks.
UR - http://www.scopus.com/inward/record.url?scp=85097784744&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85097784744&partnerID=8YFLogxK
U2 - 10.1109/SRDS51746.2020.00028
DO - 10.1109/SRDS51746.2020.00028
M3 - Conference contribution
AN - SCOPUS:85097784744
T3 - Proceedings of the IEEE Symposium on Reliable Distributed Systems
SP - 205
EP - 214
BT - Proceedings - 2020 International Symposium on Reliable Distributed Systems, SRDS 2020
PB - IEEE Computer Society
Y2 - 21 September 2020 through 24 September 2020
ER -