Comparative Performance and Communication Overhead Analysis of Post-Quantum Cryptographic Algorithms for Quantum-Safe Systems Architectures: Implications for Post-Quantum Secure Systems

Abstract

Post-quantum cryptography offers a promising safeguard for communication systems in the era of quantum-computing threats, it is essential to consider the computational efficiency and communication load of these systems when implementing them on the ground. This study compared the performance of ML-KEM and ML-DSA post-quantum algorithms to both the classical RSA-4096 and ECDSA-P256 algorithms on benchmarks such as execution time, variability, public key size, size-expansion ratio, communication overhead and signature/ciphertext size. The performance of ML-KEM variants was significantly better than RSA-4096 on the key generation, encapsulation, and decapsulation operations, while the size of the ciphertext and communication overhead were larger for the higher levels of the parameters. The advantages of ML-DSA variants were observed as their key generation was faster and the verification was comparable with ECDSA-P256 but the signing was slower and the size of the signature was much larger. The communication overhead of ML-KEM ranged between 50.00% and 206.25% over the classical counterparts, while that of ML-DSA ranged from 3,308.45% to 6,416.90% over the classical counterparts. Increasing security parameters also showed greater communication and computation costs for stronger configurations. The results highlight the need to carefully consider multiple factors beyond execution speed when selecting post-quantum algorithms for migration to quantum-safe systems, including performance efficiency, cryptographic object size, and communication requirements. These results give quantitative justification for choosing the application-specific processing capacity, bandwidth, security, constraints, and deployment conditions for the selection of standardized post-quantum mechanisms.