HIGHLIGHTS
- What: The feature enhancement module controls the degree of information retention through the gating unit, enabling the model to focus on more important feature information. To verify the effectiveness of the module proposed in this paper, ablation experiments are carried out under the signal-to-noise ratio of -4dB. It can be seen from Fig 10 that the model presented in this paper has the highest recognition accuracy under D-E conditions, which is 98.86%, which is 2.43% higher than that of MSC-RCDN and 6.15% higher than that of WKCNN.
- Who . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.