基于信源信道联合编码的遥感图像传输

摘要

A joint source and channel coding scheme based on set partitioning in hierarchical trees （SPIHT） and low density parity cheek （LDPC） codes is proposed to meet the need of high quality remote sensing image transmission. The scheme makes use of the fact that the SPIHT compressed bit stream is ordered by significance and that the bit error rate of LDPC is closely related to the code rate and iteration number of decoder to apply unequal protection to the SPIHT-compressed-stream. And thus the disadvantage of high sensitivity to noise for compressed bit stream is removed. Simulations of additive white Gaussian noise （AWGN） channel show that, with low EUNo （energy per bit over noise density ratio）, peak signal to noise ratio （PSNR） of the recovered image of the scheme is higher than that of traditional separated coding scheme by at least 7dB.%针对遥感应用对图像质量日益提高的需要,提出了一种基于层次树的集合分割算法（SPIHT）和低密度校验（LDPC）码联合编码的遥感图像高质量传输方案。此方案充分利用SPIHT编码后数据流按重要性排序,以及LDPC码的误码率性能与码率和迭代次数密切相关的特性,克服了SPIHT压缩数据流对噪声敏感的缺陷。高斯信道仿真结果表明,在码元噪谱密度比（Eb/N0）较低的情况下,利用此传输方案获得的图像恢复峰值信噪比（PSNR）,比传统的分离编码高7dB以上。