Joint Source Coding and Modulation for Low-Complexity Video Transmission
In this paper, the authors propose a Statistical QAM algorithm for video over wireless transmission - a joint approach for source coding and modulation utilizing two following ideas. Firstly context prediction is used to make distribution of image pixels values closer to bilateral geometrical distribution. Then the resultant values are mapped into the constellation points according to their frequency: more frequent values correspond to the points with lower energy level. It is shown that the proposed solution has significant power consumption gain if they consider power of video transmission and compression units jointly. The approach is analyzed from the position of bit error rate, peak-to-average power ratio, implementation costs and image quality (PSNR) at the receiver. All explanations are shown for QAM modulation system.