摘要
提出了一种基于CS(压缩感知)的LDPC码(低密度奇偶校验码)的抗误码方法。在编码端对H.264视频流中的每一个NAL(网络提取层)进行基于IEEE802.16e的LDPC编码。在解码端进行两种方法的解码,一种进行传统的LDPC解码,另一种根据压缩感知线性解码和信道码线性解码之间的联系,将基于IEEE802.16e的LDPC编解码中的校验矩阵作为压缩感知重构的测量矩阵进行压缩感知重构解码。实验结果表明,在高误码率的情况下,基于CS的LDPC码解码方法相对于传统LDPC解码方法的抗误效果更明显,为视频的抗误传输提供了新的思路。
A decoding method of LDPC ((Low Density Parity Check) codes based on CS (Compressed Sensing) is presented. Each network abstrac- tion layer (NAL) of H. 264 video stream is encoded with LDPC based on IEEE 802.16e at the encoder side. Two decoding methods at the decoder side are used. One uses LDPC decoding method. According to the connection between the compressed sensing linear programming decoding(CS-LPD) and the channel coding linear programming decoding (CC-LPD), another uses the parity-check matrix of LDPC codes in IEEE 802.16e as the measurement matrix of CS reconstruction. The experiment results show that the error-resilience effect of compressed sensing reconstruction is superior to the traditional LDPC decoding method in the situation of high bit error rate. It provides a new train of thought to the error-resilience transmission.
出处
《电视技术》
北大核心
2014年第13期112-116,共5页
Video Engineering
基金
国家自然基金面上项目(61171077)
交通部应用基础研究项目(2011-319-813-510)
南通大学创新人才基金项目(2009)