Optimal multiuser spread-spectrum data embedding in video streams

In this paper, we intend to hide multiuser data in a given host video stream with imperceptible spread-spectrum embedding. First, host video frames are picked in the original given video stream according to a frame selection pattern. We partition each host video frame into many small blocks. Based on a two-dimensional transformation of each small block and zigzag scanning, we construct the host video vectors. The embedded message for each user will be spread out with a signature and added to the host vectors. We present the orthonormal signature set of embedding carriers that achieves maximum sum signal-to-interference-plus-noise ratio (sum-SINR) at the linear filter output for any fixed embedding amplitudes. Then, for any given total embedding distortion constraint, we give the optimal multi-signature assignment and amplitude allocation pair that maximizes the sum capacity of the concealment procedure. Numerical results demonstrate the effectiveness of the proposed optimal data embedding methods in video streams.