A Reconfigured RLWE-Based Mutual Authentication Protocol for Post- Quantum IoT Security

The conventional cryptographic algorithm faces a threat due to the evolution of quantum computing algorithms. In recent days, IoT devices demand authentication techniques to ensure a perfectly secure system. Thus, to address this issue, this paper presents a Modified Ring Learning with Errors–based Mutual Authentication (MRMA) scheme tailored for resource-constrained IoT environments. The computational hardness of the RLWE problem, the proposed scheme employs polynomial-level optimizations to reduce key size and computational overhead, also preserving post-quantum security in IoT devices. To further enhance the security–efficiency trade-off, a Quantum Approximate Optimization Algorithm (QAOA) driven parameter tuning layer is introduced, formulating cryptographic parameter selection as a concern for a multi-objective optimization problem over polynomial degree, modulus, and noise distribution. Also, to improve latency and throughput, the assistance of QAOA is needed. The traditional scheme requires larger keys (512–2048 bits) and high processing time. Our performance analysis proves that the proposed MRMA scheme implements a compact 256-bit key, includes a low authentication latency of 0.0065 seconds, a throughput of 900 operations per second, and an Authentication Acceptance Rate (AAR) of 99.2%. The results prove that the performance is better than typical RLWE- and MLWE-based IoT authentication schemes. Thus, the proposed MRMA scheme supports a secure, lightweight authentication framework, making it suitable for IoT and post-quantum applications.