论文标题
转移的修剪以解码极性代码
Shifted Pruning for List Decoding of Polar Codes
论文作者
论文摘要
在连续的取消列表（SCL）解码中，树修剪操作在每个解码步骤中都保留了相对于度量的最佳路径。但是，正确的路径可能是由于施加罚款而导致的最差路径之一。在这种情况下，正确的路径被修剪并且解码过程失败。在这项工作中，我们建议在解码失败时进行其他解码尝试，其中修剪窗口不一定选择L最佳路径，但是此窗口在排序列表中的位置1和2L之间移动。在最简单的形式中，在解码步骤中选择了最坏的路径，在该步骤中，消除正确路径的可能性很高。此外，我们概括了该方案，并提出了许多变体，例如受约束的转移，嵌套的转移和分段解码下的转移，旨在降低计算复杂性。长度为512的极性代码为0.5和0.8和列表尺寸L = 2、8、32的数值结果表明，转移的方案可以提供0.25-0.5 dB的误差校正性能增益，而平均计算复杂性方法在实践范围内将常规的复杂性解码。
In successive cancellation list (SCL) decoding, the tree pruning operation retains the L best paths with respect to metric at every decoding step. However, the correct path might be among the L worst paths due to imposed penalties. In this case, the correct path is pruned and the decoding process fails. In this work, we propose a scheme for additional decoding attempts when decoding fails, in which the pruning window does not necessarily select the L best paths, but this window is shifted between positions 1 and 2L in the sorted list. In the simplest form, the L worst paths are selected at the decoding step where the probability of elimination of the correct paths is high. Additionally, we generalize the scheme and propose a number of variants such as constrained shifting, nested shifting and shifting under segmented decoding, aiming to reduce the computational complexity. The numerical results for polar codes of length 512 with code rates 0.5 and 0.8 and list sizes L=2, 8, 32, show that the shifted-pruning scheme can provide 0.25-0.5 dB gain in error correction performance, while the average computational complexity approaches the conventional list decoding complexity at practical FER ranges.
