A Novel DNA‐Algebraic Block Cipher for Satellite Communication Security Using Biological Codon Diffusion and GL(8,GF(2)) Transformations

Satellite communication systems transmit highly sensitive telemetry, command‐and‐control, and Earth‐observation data over hostile radio‐frequency channels that are exposed to eavesdropping and injection attacks. Classical block ciphers, while mathematically rigorous, do not exploit the inherent redundancy‐reduction and diffusion potential of biological models, nor are they tailored to the constraint profile of space‐borne hardware. This paper proposes a block encryption scheme that bridges DNA computing, finite‐field algebra over , and the general linear group into a unified, multi‐round cryptographic framework designed specifically for satellite communication security. The central cryptographic primitive is a DNA‐Algebraic S‐box constructed through a four‐stage pipeline: (i) byte‐to‐DNA‐codon encoding; (ii) biological strand complementation followed by codon‐level XOR using Chargaff's base‐pairing rules; (iii) multiplicative inversion in modulo the AES irreducible polynomial; and (iv) an invertible affine transformation drawn from with a satellite‐mission‐specific constant. The resulting S‐box achieves a uniform nonlinearity of across all eight output bits matching the theoretical optimum of the AES S‐box a differential uniformity of , a linear probability of , a strict avalanche criterion mean of 0.505, and zero fixed points. An expanded evaluation of fifty independently generated session S‐boxes confirms and for every session, while showing that the number of realized fixed points varies with the session strand and that the effective session‐key space is limited to 256 distinct strands, a bound that is analyzed as part of a dedicated weak‐key and equivalent‐key study. Three multi‐round block encryption algorithms of increasing complexity are constructed upon this S‐box, together with a DNA‐aware key schedule engineered for adequate related‐key diffusion. The primary scheme (Algorithm C) additionally incorporates a DNA‐MixColumns diffusion layer that operates by complementing GF‐weighted mixed bytes at the codon level, a Codon‐Diffusion function modelling biological strand‐slippage mutation, and the DNA‐aware key schedule. At 16 rounds, Algorithm C achieves a plaintext avalanche effect of and a key sensitivity of , converging to the ideal by Round 4. An ablation study isolates the individual contribution of DNA‐MixColumns and Codon‐Diffusion, showing that each layer alone is sufficient for asymptotic convergence but that removing both diffusion layers caps the avalanche effect at approximately regardless of round count, confirming that these two components rather than the byte rotation alone are the primary source of inter‐byte diffusion. The Pearson correlation between plaintext and ciphertext bytes is , statistically indistinguishable from zero. Rather than claiming unconditional superiority, the proposed scheme is presented as a competitive, session‐diversified alternative to AES, with comparative claims moderated accordingly. Resistance to a wider battery of modern cryptanalytic attacks algebraic, integral, impossible‐differential, boomerang, zero‐correlation, division‐property, invariant‐subspace, and related‐key is analyzed, and estimated hardware costs together with side‐channel and fault‐injection considerations for space‐borne deployment are discussed. The framework is validated through extensive empirical experiments and formal security arguments, establishing it as a candidate worth further study for integration into satellite on‐board cryptographic modules.