Adaptive artificial noise beamforming for securing grant-free massive machine-type communication

Massive machine-type communication (mMTC) growth beyond 5G (B5G)/6G networks presents significant security challenges, particularly in grant-free scenarios where traditional cryptographic methods are insufficient. The lack of defined access control and the complexities of key management expose these systems to vulnerabilities such as pilot contamination and jamming. Existing physical layer security (PLS) techniques, including adaptive beamforming and artificial noise generation, are limited by their static nature and reliance on perfect channel state information (CSI), making them less effective in the dynamic and densely populated environments characteristic of mMTC. This paper introduces Adaptive Artificial Noise Beamforming (AANB), an enhanced PLS approach designed to optimise the trade-off between security and system performance for grant-free mMTC in B5G/6G, by dynamically adjusting beamforming vectors and artificial noise based on real-time CSI and spatial correlation, while ensuring minimal impact on legitimate users and maximising interference against eavesdroppers. The proposed AANB’s secrecy outage probability (SOP) for grant-free mMTC is analytically derived, and the impact of AANB is demonstrated through simulations, which shows that AANB significantly lowers SOP when benchmarked against Semi-grant-free (SGF) and traditional beamforming with artificial noise (BF+AN) techniques in grant-free mMTC environment. The results indicate that AANB consistently outperforms SGF and BF+AN, achieving lower SOP values across various signal-to-noise ratio (SNR) levels and spatial correlation scenarios, offering robust security in grant-free mMTC scenarios. Additionally, bit error rate (BER) analysis demonstrates that AANB consistently outperforms benchmark schemes across all SNR levels, due to its adaptive noise-to-signal ratio optimisation, thereby enhancing resistance against eavesdropping in grant-free mMTC.