Performance and Security Trade-offs in Keyless Data Encryption and Decryption

Abstract

Traditional cryptographic systems heavily depend on secure key management, distribution, and storage, creating vulnerabilities that become critical in the face of quantum computing threats and complex network environments. This paper explores keyless data encryption and decryption paradigms that derive security from intrinsic properties of communication channels, data context, physical environments, or biological traits, eliminating the need for pre-shared or stored keys. Drawing on information-theoretic foundations such as Shannon's perfect secrecy and Wyner's wiretap channel model, the review examines key architectural modalities, including Physical Layer Security (PLS) techniques (e.g., artificial noise injection and unshared secret key generation in 5G/6G networks), the Keyless User Defined Optimal Security (KUDOS) algorithm for memory context protection, All-or-Nothing Transforms (AONT) combined with information dispersal for distributed storage, biometric-based fuzzy vault schemes, chaos-based image encryption, and emerging quantum-resilient AI-integrated frameworks like QSAFE-MM1.A detailed analysis of performance-security trade-offs reveals inherent tensions, such as the Secrecy-Reliability Performance Trade-off (SRPT) in NOMA systems, energy efficiency constraints in IoT environments, computational overhead in fully homomorphic encryption, and acceptance-rate declines in biometric systems with increasing polynomial complexity. While keyless approaches offer advantages in reduced attack surface, lower latency, and quantum resistance, they demand careful balancing against reliability, throughput, energy consumption, and scalability limitations.The findings underscore the shift toward adaptive, hybrid security architectures that integrate physical randomness, AI-driven monitoring, and context-aware mechanisms to achieve robust, future-proof protection in heterogeneous digital ecosystems.