基于多维正交载体的可裂解流指纹方案

摘要

Aimed at problems of unknown source identity, uncontrollable net-flow exchange and untraceable flow- ex-changing paths, it proposes cracking-resistance net-flow fingerprint scheme based on multi-dimensional orthogonal carri-ers. It uses two mutually orthogonal carriers so as to improve the capacity of fingerprint information. It achieves crack-ing-resistance by interval centroid carrier characteristics and hidden Markov model based decoding technique. Ultimately, it improves the robustness of scheme. What’s more, it analyzes the resistance of multi-flow attack ability by using cen-troid attribute value to select embedding carriers randomly. Besides, it analyzes accuracy of fingerprint centroid algorithm and quantization index modulation decoding technique based on HMM under different cases. Finally, its invisibility and robustness is evaluated by experiments.%针对流交换中流源身份不可知、流交换范围不可控和流路径不可追踪问题，提出了基于多维正交载体的可裂解流指纹方案。利用2种相互正交的载体提高指纹信息的容量，并通过时间间隔重心载体特性和基于隐马尔科夫模型的解码技术实现可裂解性，提高方案的健壮性。分析了基于重心属性值随机选取载体的反制多流攻击能力，以及不同条件下指纹重心标记算法和基于隐马尔科夫模型解码技术的正确率。最后通过实验对算法的健壮性和隐蔽性进行了探讨。