论文部分内容阅读
提出一种基于空间策略网格图的符号级联合信源信道编译码快速算法。构建联合译码的平面网格图,可以实现比比特级译码更好的译码性能。然而平面网格图结构复杂,使得符号级联合信源信道译码过程复杂度高。提出一种新的高效空间策略网格,通过满足限制条件方程对可变长符号的权重值进行优化,结合可变长符号后验概率译码,形成一个整体快速编译码算法。两项仿真实验的结果表明,相比平面网格图,该方法减少译码复杂度近50%,并使系统在相同功率信噪比条件下传输的比特错误率更低。这种方法可以很好地保护可变长编码后的信息数据,适用于资源受限的环境中的符号级联合信源信道编译码。
This paper proposes a fast symbol-level joint source-channel coding and decoding algorithm based on spatial strategy grids. The construction of a joint flat-panel grid decoding, decoding can be better than the bit-level decoding performance. However, the structure of the planar grid graph is complex, making the symbol-level joint source channel decoding process complicated. A new efficient spatial strategy grid is proposed. The weight value of variable length symbol is optimized by satisfying the constraint equation, and combined with the variable length symbol a posteriori probability decoding, an overall fast coding and decoding algorithm is formed. The results of two simulation experiments show that this method can reduce the decoding complexity by nearly 50% and make the system transmit the bit error rate under the same power SNR even lower than the planar grid graph. This method can well protect the variable-length encoded information data and is suitable for symbol-level joint source channel coding and decoding in resource-limited environments.