Neural Cryptography Based on Generalized Tree Parity Machine for Real-Life Systems

Abstract

Traditional public key exchange protocols are based on algebraic number theory. In another perspective, neural cryptography, which is based on neural networks, has been emerging. It has been reported that two parties can exchange secret key pairs with the synchronization phenomenon in neural networks. Although there are various models of neural cryptography, called Tree Parity Machine (TPM), many of them are not suitable for practical use, considering efficiency and security. In this paper, we propose a Vector-Valued Tree Parity Machine (VVTPM), which is a generalized architecture of TPM models and can be more efficient and secure for real-life systems. In terms of efficiency and security, we show that the synchronization time of the VVTPM has the same order as the basic TPM model, and it can be more secure than previous results with the same synaptic depth.

Chaos-guided neural key coordination for improving security of critical energy infrastructures

“…Table 3 presents the results of the NIST test suite [16] on the synchronized neural session In the synchronized neural key, the outcome of the frequency test denotes a proportion of 1 and 0. The frequency test outcome of the TLVVNN technique is 0.711478, that is better than the outcome 0.538632 in Jeong et al [6], 0.512374 in n Dong and Huang [3], 0.1329 in Karakaya et al [8], 0.632558 in Patidar et al [19], and 0.629806 in Liu et al [9]. The frequency test's P Value comparison is presented in Table 4.…”

“…Huang [3], Jeong et al [6], and Teodoro et al [30] were first looked at in this study. Their shortcomings were also mentioned in this study.…”

“…Although they only examined only the geometric attack, it is uncertain whether the CVTPM can withstand a majority attack. Jeong et al [6] have proposed a VVTPM scheme. However, this method does not provide an effective syncing evaluation.…”

Nature-Inspired Gravitational Search-Guided Artificial Neural Key Exchange for IoT Security Enhancement

“…It is observed that, for the large size of the network, Random Walk's rule outweigh the other two learning rules. The TLTPM and VVTPM [25] approaches for synchronization time using the Hebbian learning are compared in table 8. The synchronization time of the TLTPM using the Hebbian learning rule is much faster than the existing VVTPM, as seen in this table.…”

“…Likelihood of Success of Attacker UsingGeometric Attack α1 − α2 − α3 − β − γ =[25, 30], γ = 3, and Hebbian learning algorithm was discussed. The value of the α1 − α2 − α3 parameter significantly affects the number of iterations in the synchronization process (the figures show three values: α1 = α2 = α3 = 6, 8, 12).…”