Hybrid Advanced Encryption Standard–Counter Mode and Adaptive Least Significant Bit Audio Steganography Framework for Secure Military Communication

Abstract

This study proposes a hybrid audio steganography framework that combines Advanced Encryption Standard in Counter Mode (AES-CTR) encryption, Reed–Solomon error correction, and a temporal masking–guided adaptive Least Significant Bit (LSB) embedding technique to enhance the security of military communications. A dataset of 60 uncompressed WAV audio files (44.1 kHz, 16-bit) was used to evaluate imperceptibility, robustness, and payload fidelity. The system embeds AES-CTR–encrypted tokens into perceptually masked audio samples to reduce detectability. Experimental results across payload sizes ranging from 10 to 50,000 bytes show that the proposed method achieves a mean Peak Signal-to-Noise Ratio (PSNR) improvement of 3.7–5.2 dB over conventional LSB and 2.1–3.4 dB over AES-LSB baselines. Bit-Error-Rate (BER) analysis demonstrates a 68% reduction in BER under additive noise after RS decoding. Performance was further assessed using SNR, payload capacity, and χ²-based steganalysis metrics, where the hybrid model consistently outperformed baseline methods. The proposed framework, therefore, provides a multilayer enhancement in confidentiality, robustness, and perceptual transparency for covert military communication. Limitations include computational overhead and reliance on simulated audio conditions, suggesting future work should explore real-time deployment and codec-resilient embedding.