Blind plain-text extraction of chaos masking-based optical chaos communications
摘要
This paper investigates the security of optical chaotic communication (OCC) based on chaos masking. A neural network-based attack model was constructed to directly extract message signals embedded in optical chaotic carriers. Using this model, the effectiveness of attacks at different masking coefficients can be quantified as the bit error rate of illegal decryption. We conducted attack tests on two main types of OCC systems and visualized the attack effects in the scenario of secure transmission of images. The results demonstrate the effectiveness of the proposed attack scheme. We believe that it can serve as a potential security testing tool for OCC systems. This is meaningful for increasing confidence in the use of the OCC.