Improved Recurrent Neural Network based BP Decoding Algorithm for Polar Codes

Table 1. [in Chinese]
View table
View in Article
Table 1. [in Chinese]
参数选项值
测试平台 TensorFlow
调制方式 BPSK
SNR范围/dB 1~5
训练集大小 7.5×10⁴
测试集大小 3×10³
优化算法 AdaGrad
损失函交叉熵
数学习率 0.001
Epochs 25

Table 2. [in Chinese]

View table

View in Article

Table 2. [in Chinese]

算法	乘法运算	加法运算	存储空间
BP	0	2T_BPNlog₂N~30720(100%)	0
DNN-BP	2TNlog₂N(100%)	2TNlog₂N~3840(12.5%)	2TNlog₂N~3840(100%)
RNN-OMS-BP	0	4TNlog₂N~7680(25%)	2N(log₂N +1)~896(23.3%)
RNN-OMS-BP-L	0	3TNlog₂N~5760(18.75%)	2N(log₂N+1)-N/8~888(23.1%)

Tools

Get Citation

Copy Citation Text

Xue-lu DENG, Da-qin PENG. Improved Recurrent Neural Network based BP Decoding Algorithm for Polar Codes[J]. Study On Optical Communications, 2022, 48(4): 17

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites