基于自适应网格编码QAM的渐进图像编码传输方案  

Transmission scheme for of progressive coded images based on adaptive trellis coded QAM

作　　者：赵锴[1] 崔慧娟[1] 唐昆[1] 

机构地区：[1]清华大学电子工程系,清华信息科学与技术国家实验室,北京100084

出　　处：《清华大学学报（自然科学版）》2012年第5期607-611,共5页Journal of Tsinghua University(Science and Technology)

基　　金：国家自然科学基金面上项目(60572081)

摘　　要：为了在无线信道下鲁棒和高效地传输静态渐进图像,基于高效率编码调制系统提出了一种新型编码传输框架。使用支持多种编码传输模式的自适应网格编码正交振幅调制作为脆弱嵌入式压缩渐进码流的编码保护方案,在给定的功率和延迟约束条件下,基于启发式并行搜索算法对待传信源压缩分组的传输功率和编码调制模式进行联合优化,使得图像传输的端对端期望失真最小。经过对不同测试图像在不同功率和延迟约束条件下的仿真,该传输优化方案相对于以往基于不等重功率分配算法的传输优化方案取得了明显性能增益,在Gauss白噪声信道和准静态衰落信道中,接收端重建图像的峰值信噪比分别平均提高了1.33和2.70dB。An efficient, robust transmission scheme is presented for progressive coded images over wireless channels using power efficient coded modulation methods. The adaptive coded trellis quadrature amplitude modulation （QAM） scheme which can support multiple coded modulation modes, is used to protect the embedded compression stream against erroneous channels. This is used with a parallel heuristic method to optimize the transmission power with coded modulation modes for coded compression packets to minimize the expected end to end transmission distortion. Extensive tests show that for various test images with various power and delay constraints, the scheme outperforms previous schemes based only on unequal power allocation methods. The average peak signal to noise ratio （PSNR） gain of reconstructed images over AWGN is 1.33 dB and over quasi static fading channels is 2.70 dB.

关 键 词：渐进图像编码 自适应网格编码正交振幅调制(QAM) 不等重功率分配 差分演化算法 

分 类 号：TN919.81[电子电信—通信与信息系统]