Enhancing the Strict Avalanche Criteria for Speech Encryption through Systematic Analysis

Abstract

By guaranteeing that a single input bit flip substantially changes the cipher
text, this paper propose a unique cryptographic architecture to improve the Strict
Avalanche Criterion (SAC) in speech ciphers and strengthen security against cryptanalytic attacks. By using XOR operations to cumulatively mix plaintext segments
with various secret keys, the method introduces non-linearity to produce a uniform bit distribution. The primary benefit is in its capacity to generate a balanced
avalanche effect with a lower computational complexity, beyond the constraints of
conventional techniques. The effectiveness of the approach was assessed through
random secret keys and mixing levels of 2, 4, and 8 quantifying performance using
standard cryptographic analysis. With a bit change rate of 49.93% for mixing level
2, increasing to 50.035% for mixing level 4, and stabilizing at 50.04% for mixing level
8, the results show a nearly optimal SAC, verifying scalability and efficiency. These
results show that by attaining robust SAC with consistent randomization, the suggested cumulative XOR technique greatly improves voice encryption systems. The
strategy is expected to help secure communication applications and offers a promising direction for future study into simple, scalable cryptographic solution and its
performance.