Security of Spectrum Learning in Cognitive Radios
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Due to delay and energy constraints, a cognitive radio may not be able to perform spectrum sensing in all available channels. Therefore, a sensing policy is needed to decide which channels to sense. The channel selection problem is the problem of designing such a sensing policy to maximize throughput while avoiding interference to primary users. The channel selection problem can be formulated as a reinforcement learning problem. Channel selection schemes that employ reinforcement machine learning algorithms are vulnerable to belief manipulation attacks that contaminate the knowledge base of the learning algorithms. In this paper, we analyze the security of channel selection algorithms that are based on reinforcement learning and propose mitigation techniques that make these algorithms more robust against belief manipulation attacks.
Abstract
Due to delay and energy constraints, a cognitive radio may not be able to perform spectrum sensing in all available channels. Therefore, a sensing policy is needed to decide which channels to sense. The channel selection problem is the problem of designing such a sensing policy to maximize throughput while avoiding interference to primary users. The channel selection problem can be formulated as a reinforcement learning problem. Channel selection schemes that employ reinforcement machine learning algorithms are vulnerable to belief manipulation attacks that contaminate the knowledge base of the learning algorithms. In this paper, we analyze the security of channel selection algorithms that are based on reinforcement learning and propose mitigation techniques that make these algorithms more robust against belief manipulation attacks.
참고문헌 (22)
- S.H. Ahmad, "Multi-channel opportunistic access: a case of restless bandits with multiple plays", proceedings of Allerton Conference on Communication, Control, and Computing (2009)
- S.H. Ahmad, "Optimality of Myopic Sensing in Multi-Channel Opportunistic Access", IEEE Transactions on Information Theory, 55(9), pp.4040~4050 (2009)
- Security of Spectrum Learning in Cognitive Radios, "B. Bahrak", Dept. of Electrical and Computer Engineering, ~, pp.Technical Report TR-ECE-12 -09 (2012)
- N. Baldo, "Learning and adaptation in cognitive radios using neural networks", 1st IEEE workshop on cognitive radio networks (in conjunction with IEEE CCNC 2008) (2008)
- M. Barreno, "The security of machine learning", Machine Learning Journal (MLJ) Special Issue on Machine Learning in Adversarial Environments, 81(2) (2008)
- M. Barreno, "Can machine learning be secure?", ACM Symposium on Information, Computer and Communication Security, pp.16~25 (2006)
- S. Boyd, "Convex Optimization" (2004), Cambridge University Press
- Y. Chen, "Joint design and separation principle for opportunistic spectrum access in the presence of sensing errors", IEEE Transactions on Information Theory, 54(5) (2008)
- T. Clancy, "Robust Signal Classification Using Unsupervised Learning", IEEE Transaction on Wireless Communication (2011)
- E.N. Gilbert, "Capacity of burst-noise channels", Bell Syst. Tech. J., 39, pp.1253~1265 (1960)
- L. Lai, "Cognitive Medium Access: Exploration, Exploitation and Competition", IEEE Transaction on Mobile Computing, 10(2), pp.239~253 (2011)
- M. Li, "Optimal Jamming Attack Strategies and Network Defense Policies in Wireless Sensor Networks", IEEE Transactions on Mobile Computing, 9(8), pp.1119~1133 (2010)
- K. Liu, "A restless bandit formulation of opportunistic access: indexablity and index policy", Proc. of the 5th IEEE Conference on Sensor, Mesh and Ad Hoc Communications and Networks Workshops (2008)
- D. Lowd, "Adversarial learning", Proceedings of the Eleventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.641~647 (2005)
- B. Nelson, "Exploiting machine learning to subvert your spam filter", Proceedings of the First USENIX Workshop on Large- Scale Exploits and Emergent Threats (2008)
- T. Newman, "Security threats to cognitive radio signal classifiers", Virginia Tech Wireless Personal Communications Symposium (2009)
- M. Ohya, "An information-theoretical approach and its application to optimal problems", Electronics and Communication in Japan, Part 3, 76(6) (1993)
- R. Smallwood, "The optimal control of partially observable Markov processes over a finite horizon", Operations Research, pp.1071~1088 (1971)
- R. S. Sutton, "Reinforcement Learning: An Introduction" (1998), MIT press
- K. Thomas, "Design and Evaluation of a Real-Time URL Spam Filtering Se", IEEE Symposium on Security and Privacy 2011 (2011)
- A. Wald, "Sequential Analysis" (1947), Wiley
- Q. Zhao, "On myopic sensing for multi- channel opportunistic access: structure, optimality, and performance", IEEE Transactions on Wireless Communications (2008)