QRAND-AVS: A Quantum-Randomness Assisted, AI-Optimized DNA–ECC Framework for Secure Video Steganography in IoT

The swift expansion of Internet of Things (IoT) applications in telemedicine, defense communication, and copyright protection has heightened the necessity for scalable, lightweight, and quantum-resistant multimedia security frameworks. AES and DES are two examples of old symmetric algorithms that are hard to use and have security holes when it comes to sharing keys. Quantum adversaries can still assault classical public-key techniques. This paper presents QRAND-AVS, a hybrid video steganography framework that combines quantum randomness, lightweight cryptography, and adaptive intelligence. The Koblitz method makes plain text messages readable, while Elliptic Curve Cryptography (ECC) makes them unreadable. To make sure that the private keys are truly random, a Quantum Random Number Generator (QRNG) is used. The ciphertext is encoded into DNA nucleotides and adaptively embedded into key video frames selected via histogram-variance analysis. Strong transform-domain embedding is possible with a two-level Discrete Wavelet Transform (DWT) and Singular Value Decomposition (SVD). An AI-based optimizer changes the rules for DNA, the strength of the embedding, and the levels of quantization in real time to find the right balance between payload and invisibility. When tested on a number of benchmark videos, the proposed method had a PSNR that was up to 2.4 dB higher and an embedding capacity that was 40% higher than the LSB and baseline SVD methods. It also cut down on encryption overhead by 35%. Security analysis shows that the system is safe from brute-force, replay, and steganalysis attacks as long as the PSNR/SSIM limits are followed. The results show that QRAND-AVS is a smart, light, and quantum-safe way to protect multimedia that works well in IoT and post-quantum communication settings.