Algorithm 2 CNN-based sensor nodes’ authentication.

Input: The $i^{\mathrm{th}}$ CSI to authenticate ${\mathit{x}}^{\left(i\right)}$

Output: The label of unknown CSI ${\widehat{\mathit{y}}}^{\left(i\right)}$, the new weight matrix ${\mathit{W}}^{◊}$, and threshold vector ${\mathit{\xi }}^{◊}$ of CNN

1: Initialize all connection weights ${\mathit{W}}_0^{◊}$, and thresholds ${\mathit{\xi }}_0^{◊}$ in the network will be obtained through the training of CNN, using the pre-acquired dataset ${\mathit{D}}^{◊}={\left\{\left({\mathit{x}}_k,{\mathit{y}}_k\right)\right\}}_{k=1}^m$;

2: Compute ${\widehat{\mathit{y}}}^{\left(i\right)}$ by the well-trained CNN;

3: Update the training set ${\mathit{D}}^{◊}={\left\{\left({\mathit{x}}_k,{\mathit{y}}_k\right)\right\}}_{k=1}^m$ by $\left({\mathit{x}}^{\left(i\right)},{\widehat{\mathit{y}}}^{\left(i\right)}\right)$;

4: Retrain the CNN by the new dataset and get new weight matrix ${\mathit{W}}^{◊}$ and threshold vector ${\mathit{\xi }}^{◊}$;

5: Return ${\widehat{\mathit{y}}}^{\left(i\right)}$, ${\mathit{W}}^{◊}$, ${\mathit{\xi }}^{◊}$.