Mobile wireless sensor networks (WSNs) are facing threats from malicious nodes that disturb packet transmissions, leading to poor mobile WSN performance. Existing studies have proposed a number of methods, such as decision tree-based classification methods and reputation based methods, to detect these malicious nodes. These methods assume that the malicious nodes follow only pre-defined attack models and have no learning ability. However, this underestimation of the capability of malicious node is inappropriate due to recent rapid progresses in machine learning technologies. In this study, we design reinforcement learning-based malicious nodes, and define a novel observation space and sparse reward function for the reinforcement learning. We also design an adaptive learning method to detect these smart malicious nodes. We construct a robust classifier, which is frequently updated, to detect these smart malicious nodes. Extensive experiments show that, in contrast to existing attack models, the developed malicious nodes can degrade network performance without being detected. We also investigate the performance of our detection method, and confirm that the method significantly outperforms the state-of-the-art methods in terms of detection accuracy and false detection rate.
Boqi GAO
Osaka University
Takuya MAEKAWA
Osaka University
Daichi AMAGATA
Osaka University
Takahiro HARA
Osaka University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Boqi GAO, Takuya MAEKAWA, Daichi AMAGATA, Takahiro HARA, "Detecting Reinforcement Learning-Based Grey Hole Attack in Mobile Wireless Sensor Networks" in IEICE TRANSACTIONS on Communications,
vol. E103-B, no. 5, pp. 504-516, May 2020, doi: 10.1587/transcom.2019EBP3151.
Abstract: Mobile wireless sensor networks (WSNs) are facing threats from malicious nodes that disturb packet transmissions, leading to poor mobile WSN performance. Existing studies have proposed a number of methods, such as decision tree-based classification methods and reputation based methods, to detect these malicious nodes. These methods assume that the malicious nodes follow only pre-defined attack models and have no learning ability. However, this underestimation of the capability of malicious node is inappropriate due to recent rapid progresses in machine learning technologies. In this study, we design reinforcement learning-based malicious nodes, and define a novel observation space and sparse reward function for the reinforcement learning. We also design an adaptive learning method to detect these smart malicious nodes. We construct a robust classifier, which is frequently updated, to detect these smart malicious nodes. Extensive experiments show that, in contrast to existing attack models, the developed malicious nodes can degrade network performance without being detected. We also investigate the performance of our detection method, and confirm that the method significantly outperforms the state-of-the-art methods in terms of detection accuracy and false detection rate.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.2019EBP3151/_p
Copy
@ARTICLE{e103-b_5_504,
author={Boqi GAO, Takuya MAEKAWA, Daichi AMAGATA, Takahiro HARA, },
journal={IEICE TRANSACTIONS on Communications},
title={Detecting Reinforcement Learning-Based Grey Hole Attack in Mobile Wireless Sensor Networks},
year={2020},
volume={E103-B},
number={5},
pages={504-516},
abstract={Mobile wireless sensor networks (WSNs) are facing threats from malicious nodes that disturb packet transmissions, leading to poor mobile WSN performance. Existing studies have proposed a number of methods, such as decision tree-based classification methods and reputation based methods, to detect these malicious nodes. These methods assume that the malicious nodes follow only pre-defined attack models and have no learning ability. However, this underestimation of the capability of malicious node is inappropriate due to recent rapid progresses in machine learning technologies. In this study, we design reinforcement learning-based malicious nodes, and define a novel observation space and sparse reward function for the reinforcement learning. We also design an adaptive learning method to detect these smart malicious nodes. We construct a robust classifier, which is frequently updated, to detect these smart malicious nodes. Extensive experiments show that, in contrast to existing attack models, the developed malicious nodes can degrade network performance without being detected. We also investigate the performance of our detection method, and confirm that the method significantly outperforms the state-of-the-art methods in terms of detection accuracy and false detection rate.},
keywords={},
doi={10.1587/transcom.2019EBP3151},
ISSN={1745-1345},
month={May},}
Copy
TY - JOUR
TI - Detecting Reinforcement Learning-Based Grey Hole Attack in Mobile Wireless Sensor Networks
T2 - IEICE TRANSACTIONS on Communications
SP - 504
EP - 516
AU - Boqi GAO
AU - Takuya MAEKAWA
AU - Daichi AMAGATA
AU - Takahiro HARA
PY - 2020
DO - 10.1587/transcom.2019EBP3151
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E103-B
IS - 5
JA - IEICE TRANSACTIONS on Communications
Y1 - May 2020
AB - Mobile wireless sensor networks (WSNs) are facing threats from malicious nodes that disturb packet transmissions, leading to poor mobile WSN performance. Existing studies have proposed a number of methods, such as decision tree-based classification methods and reputation based methods, to detect these malicious nodes. These methods assume that the malicious nodes follow only pre-defined attack models and have no learning ability. However, this underestimation of the capability of malicious node is inappropriate due to recent rapid progresses in machine learning technologies. In this study, we design reinforcement learning-based malicious nodes, and define a novel observation space and sparse reward function for the reinforcement learning. We also design an adaptive learning method to detect these smart malicious nodes. We construct a robust classifier, which is frequently updated, to detect these smart malicious nodes. Extensive experiments show that, in contrast to existing attack models, the developed malicious nodes can degrade network performance without being detected. We also investigate the performance of our detection method, and confirm that the method significantly outperforms the state-of-the-art methods in terms of detection accuracy and false detection rate.
ER -