Optimum Bit Rate for Image Transmission over Underwater Acoustic Channel

Abstract

In this paper, image transmission in underwater channels is considered. The images are encoded with forward error correction using an unequal error protection technique together with the Reed-Solomon codes and dynamic bit-rate allocation before transmitted. This paper proposes a novel rate allocation scheme for efficient image bit stream transmission in underwater acoustic channels with optimum bit rates. The optimality is achieved in the sense that the comprehensive peak signal–to–noise ratio of the image transmission is maximized under channel bit rate and bit error rate constraints. Based on a modified set partitioning in hierarchical trees (M-SPIHT) image coder, four different flocks of bit-streams based on their significance levels are generated. According to their significance levels, the blocks of the significant bits, the sign bits, the set bits and the refinement bits are transmitted with different protection levels, so as to reduce the total distortion of received image. In addition to the careful selection of each component and intuitive justification in the detailed system design, simulation results have also been included. It is demonstrated that the proposed scheme outperforms the equal error protection for image transmission in underwater channels, significantly improves the peak signals–to–noise ratio (PSNR) performance in comparison to existing coding schemes.