Algorithm 2 The proposed KLD technique for a CR-IoT network.

Input: $M,N_s,f_s,T,{\tau }_s,$ and ${\tau }_r$ Output: Calculate the probability of a false alarm $p_{f,FC}^{KLD},$ the probability of detection $p_{d,FC}^{KLD},$ and the sum rate $R^{KLD}$ 1:Initialize $N_s,M$   2:fori from M do 3: Calculate: $e_i={\sum }_{n=1}^{N_s}{\left|z_i\left(\frac{n}{f_s}\right)\right|}^2$   4: Set: $e_i\sim \left\{\begin{array}{l}\aleph \left({\mu }_{0,i}\left(H_0\right),{\sigma }_{0,i}^2\left(H_0\right)\right)\\ \aleph \left({\mu }_{1,i}\left(H_1\right),{\sigma }_{1,i}^2\left(H_1\right)\right)\end{array}\right.$   5: Calculate: ${\overline{\mu }}_{0,i}=\frac{{\sum }_{i=1}^M{\mu }_{0,i}-{\mu }_{0,i}}{(M-1)}$ and ${\overline{\mu }}_{1,i}=\frac{{\sum }_{i=1}^M{\mu }_{1,i}-{\mu }_{1,i}}{(M-1)}$ with ${\mu }_{0,i}$ and ${\mu }_{1,i}$ in Equation (13)   6: Calculate: ${\overline{\sigma }}_{0,i}^2=\frac{{\sum }_{i=1}^M{\sigma }_{0,i}^2-{\sigma }_{0,i}^2}{(M-1)}$ and ${\overline{\sigma }}_{1,i}^2=\frac{{\sum }_{i=1}^M{\sigma }_{1,i}^2-{\sigma }_{1,i}^2}{(M-1)}$ with ${\sigma }_{0,i}^2$ and ${\sigma }_{1,i}^2$ in Equation (14)   7: Calculate: $KLD\left(i\right)=\frac{1}{2}\left[log\left(\frac{{\overline{\sigma }}_{0,i}^2}{{\overline{\sigma }}_{1,i}^2}\right)-1+\left(\frac{{\overline{\sigma }}_{1,i}^2}{{\overline{\sigma }}_{0,i}^2}\right)+\frac{{\left({\overline{\mu }}_{1,i}-{\overline{\mu }}_{0,i}\right)}^2}{{\overline{\sigma }}_{0,i}^2}\right]$   8:end for  9:if${\sum }_{i=1}^{M}KLD\left(i\right)<{\beta }_{KLD}$then 10: Set: $p_{f,FC}^{KLD}=1$   11:else 12: Set: $p_{f,FC}^{KLD}=0$   13:end if  14:if${\sum }_{i=1}^{M}KLD\left(i\right)\ge {\beta }_{KLD}$then 15: Set: $p_{d,FC}^{KLD}=1$   16:else 17: Set: $p_{d,FC}^{KLD}=0$   18:end if  19:Calculate: $R^{KLD}=\alpha p_{d,FC}^{KLD}{R}_{PU}+\left(1-\alpha \right)\left(1-p_{f,FC}^{KLD}\right)R_{CR-IoT}$