Neural network-based symmetric encryption algorithm with encrypted traffic protocol identification

Cryptography is a cornerstone of power grid security, with the symmetry and asymmetry of cryptographic algorithms directly influencing the resilience of power systems against cyberattacks. Cryptographic algorithm identification, a critical component of cryptanalysis, is pivotal to assessing algorithm security and hinges on the core characteristics of symmetric and asymmetric encryption methods. A key challenge lies in discerning subtle spatial distribution patterns within ciphertext data to infer the underlying cryptographic algorithms, which is essential for ensuring the communication security of power systems. In this study, we first introduce a plaintext guessing model (SCGM model) based on symmetric encryption algorithms, leveraging the strengths of convolutional neural networks to evaluate the plaintext guessing capabilities of four symmetric encryption algorithms. This model is assessed for its learning efficacy and practical applicability. We investigate protocol identification for encrypted traffic data, proposing a novel scheme that integrates temporal and spatial features. Special emphasis is placed on the performance of algorithms within both symmetric and asymmetric frameworks. Experimental results demonstrate the effectiveness of our proposed scheme, highlighting its potential for enhancing power grid security.