Skip to main content


Springer Nature is making Coronavirus research free. View research | View latest news | Sign up for updates

Table 1 Procedures of batch normalization

From: A robust modulation classification method using convolutional neural networks

Input: B, γ, β
Output: y={y1,y2,...,ym}
1. Calculate the mean and variance value of B:
  \(\mu _{B}=\frac {1}{m}\sum _{i=1}^{m}x_{i},\)
  \(\sigma _{B}^{2}=\frac {1}{m}\sum _{i=1}^{m}(x_{i}-\mu _{B})^{2}\)
2. Normalize
  \(\hat {x_{i}}=\frac {x_{i}-\mu _{B}}{\sqrt {\sigma _{B}^{2}+\epsilon }}\)
3. Scale and shift
  \(y_{i}=\gamma \hat {x_{i}}+\beta \)