An adaptive physical-layer security enhancement framework for ambient backscatter communication in industrial IoT

Ambient backscatter communication (AmBC) has emerged as a promising ultra-low power consumption communication paradigm for industrial Internet of Things (IIoT), leveraging existing radio frequency (RF) sources such as 4G/5G base stations to transmit industrial data without requiring additional power consumption. However, the inherently passive and open nature of AmBC makes it vulnerable to eavesdropping attacks from unauthorized devices, posing severe security and reliability threats to sensitive industrial data. In this paper, we propose an adaptive physical-layer security enhancement (APSE) framework that dynamically integrates multiple physical-layer security (PLS) techniques, including artificial noise (AN), beamforming (BF), antenna selection (AS), intelligent reflecting surface (IRS), and relay-assisted schemes. By leveraging real-time environmental sensing and reinforcement learning-based adaptive decision-making, the proposed APSE framework continuously monitors channel state information and threat scenarios, intelligently selecting optimal PLS technique combinations to minimize secrecy outage probability (SOP) and bit error rate (BER). Moreover, generalized spatial modulation (GSM), a physical layer modulation technique utilizing spatial diversity, is proposed to assist the APSE framework (APSE-GSM) to further improve the SOP and BER performance. Extensive simulations under realistic industrial channel conditions demonstrate that the proposed APSE framework significantly outperforms traditional static and single-technique schemes. In particular, integrating the GSM scheme further enhances security and reliability, achieving the lowest SOP and BER with manageable energy consumption. The results highlight the potential of the proposed APSE framework for adaptive, robust, and energy-efficient security solutions in dynamic IIoT environments.