Robust Multimodal Steganography Framework for Secure Transmission of Text, Image, and Audio Data
The growth of development the digital communication has greatly raised the chances of information leakage, interception, and unauthenticated access, particularly in heterogeneous multimedia setting. Conventional single-modal steganographic methods have been known to be limited in their payload storage capacity, weak, and susceptible to steganalysis attacks. In order to overcome these issues, the proposed work develops a Robust Multimodal Steganography Framework to ensure the secure transmission of the text, image, and audio information over the untrusted communication channels. The suggested framework combines the utilization of adaptive embedding plans in various carrier media with the Discrete Wavelet Transform (DWT) being employed in image stealth, the Discrete Cosine Transform (DCT) being employed in audio embedding and AES-based encryption being employed in protecting the textual payload. An active payload allocation scheme provides maximum imperceptibility and resistance to noise, compression and signal distortions. The experimental analysis of the proposed framework shows that this framework has better security and robustness than traditional unimodal methods. With the system, attains an average PSNR of 52.6 dB with stego-images, SNR of 41.8 dB with stego-audio and a bit error rate of less than 0.7 with common attack conditions like Gaussian noise and compression. It has an enhanced payload capacity of up to 32 and the transparency to perceptions is still high. The framework also has high resistance to statistical steganalysis and the detection accuracy dwindles to less than 18 %.