Deep-learning-based ciphertext-only attack on optical double random phase encryption

作     者：Meihua Liao Shanshan Zheng Shuixin Pan Dajiang Lu Wenqi He Guohai Situ Xiang Peng Meihua Liao;Shanshan Zheng;Shuixin Pan;Dajiang Lu;Wenqi He;Guohai Situ;Xiang Peng

作者机构：Key Laboratory of Optoelectronic Devices and System of Ministry of Education and Guangdong ProvinceCollege of Physics and Optoelectronic EngineeringShenzhen UniversityShenzhen 518060China Shanghai Institute of Optics and Fine MechanicsChinese Academy of SciencesShanghai 201800China Center of Materials Science and Optoelectronics EngineeringUniversity of Chinese Academy of SciencesBeijing 100049China Hangzhou Institute for Advanced StudyUniversity of Chinese Academy of SciencesHangzhou 310000China. 

出 版 物：《Opto-Electronic Advances》 (光电进展（英文）)

年 卷 期：2021年第4卷第5期

页      面：12-23页

核心收录：

学科分类：11[军事学] 12[管理学] 1105[军事学-军队指挥学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 0839[工学-网络空间安全] 081104[工学-模式识别与智能系统] 08[工学] 110505[军事学-密码学] 0835[工学-软件工程] 110503[军事学-军事通信学] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：financial supports from the National Natural Science Foundation of China(NSFC)(62061136005,61705141,61805152,61875129,61701321) Sino-German Research Collaboration Group(GZ 1391) the Mobility program(M-0044)sponsored by the Sino-German Center Chinese Academy of Sciences(QYZDB-SSW-JSC002) Science and Technology Innovation Commission of Shenzhen(JCYJ20170817095047279) 

主　　题：optical encryption random phase encoding ciphertext-only attack deep learning 

摘      要：Optical cryptanalysis is essential to the further investigation of more secure optical ***-based at-tack of optical encryption eliminates the need for the retrieval of random phase keys of optical encryption systems but it is limited for practical applications since it requires a large set of plaintext-ciphertext pairs for the cryptosystem to be ***,we propose a two-step deep learning strategy for ciphertext-only attack(COA)on the classical double ran-dom phase encryption(DRPE).Specifically,we construct a virtual DRPE system to gather the training ***,we divide the inverse problem in COA into two more specific inverse problems and employ two deep neural networks(DNNs)to respectively learn the removal of speckle noise in the autocorrelation domain and the de-correlation operation to retrieve the plaintext *** these two trained DNNs at hand,we show that the plaintext can be predicted in real-time from an unknown ciphertext *** proposed learning-based COA method dispenses with not only the retrieval of random phase keys but also the invasive data acquisition of plaintext-ciphertext pairs in the DPRE *** simulations and optical experiments demonstrate the feasibility and effectiveness of the proposed learning-based COA method.