A Method for Detecting LDoS Attacks in SDWSN Based on Compressed Hilbert–Huang Transform and Convolutional Neural Networks

. 2023 May 14;23(10):4745. doi: 10.3390/s23104745

Algorithm 1 Traffic feature information extraction.

Require: Network traffic data $f l o w_t$

Ensure: Feature set Q

1: for $\forall (S r c_{ip}, D e s_{ip}, P_{t}, C_{t}) \in f l o w_t$ do

2: for each time interval t do

3: if $S r c_{ip} \to \exists$ then

4: $(S r c_{ip}, c o u n t) \leftarrow c o u n t + 1$

5: else

6: $(S r c_{ip}, c o u n t) \leftarrow 1$

7: end if

8: if $D e s_{ip} \to \exists$ then

9: $(D e s_{ip}, c o u n t) \leftarrow c o u n t + 1$

10: else

11: $(D e s_{ip}, c o u n t) \leftarrow 1$

12: end if

13: $\frac{P_{t + 1} - P_{t}}{P_{t}} \leftarrow D P T R$

14: $\frac{C_{t + 1} - C_{t}}{C_{t}} \leftarrow D C C R$

15: $(S r c_{ip}, c o u n t) \leftarrow U S I P$

16: $\frac{(D e s_{ip}, c o u n t)}{P_{t}} \leftarrow N U D I P$

17: end for

18: end for