Abstract—Wireless communication systems are increasingly vulnerable to privacy and security threats, necessitating efficient and adaptive protection mechanisms. This paper aims to enhance wireless security by proposing an artificial noise-based framework integrated with machine learning for dynamic optimization of noise at the physical layer. The methodology involves real-time monitoring of network conditions, prediction of optimal noise levels using a machine learning model, and adaptive noise injection to degrade the signal quality at the eavesdropper while preserving reliable communication for legitimate users. A mathematical model is employed to evaluate the effectiveness of the approach using key performance metrics such as secrecy capacity, Bit Error Rate (BER), Signal-to-Noise Ratio (SNR), and system throughput. Simulation and empirical results demonstrate that the proposed approach improves secrecy capacity up to 0.90 bits/s/Hz while maintaining low BER and stable throughput without compromising bandwidth or power efficiency. The integration of artificial noise with machine learning enables enhanced adaptability and robustness across diverse network scenarios. This work contributes a scalable and efficient framework for improving wireless communication privacy, providing a practical solution for next-generation secure communication systems.

Index Terms—adaptive security protocols, artificial noise, physical layer, security protocols, signal obfuscation, wireless privacy

Cite: Santosh Kumar S., Keshavamurthy S., and Sunil Kumar K. N., "Adaptive Artificial Noise and Machine  Learning-based Framework for Secure Wireless Communication," International Journal of Electrical and Electronic Engineering & Telecommunications, vol. 15, no. 5, pp. 323-333, 2026. doi: 10.18178/ijeetc.15.5.323-333

Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).