Deep-Learning-Based Source Reconstruction Method Using Deep Convolutional Conditional Generative Adversarial Network

Keyphrases

• Deep Learning 100%
• Depthwise Separable Convolution 100%
• Source Reconstruction 100%
• Conditional Generative Adversarial Network (cGAN) 100%
• Equivalent Sources 85%
• Scattered Field 42%
• Scatterer 28%
• Computational Approach 14%
• New Paths 14%
• Numerical Examples 14%
• Low Computational Complexity 14%
• Deep Convolutional Neural Network (deep CNN) 14%
• High Computational Cost 14%
• Transmitter 14%
• Ill-conditioned Problem 14%
• Far-field Measurement 14%
• Electromagnetic Computation 14%
• Online Training 14%
• Forward Simulator 14%

Engineering

• Deep Learning Method 100%
• Scatterer 28%
• Field Data 28%
• Discriminator 28%
• Measured Scattered Field 28%
• Numerical Example 14%
• Computational Cost 14%
• Convolutional Neural Network 14%
• Conventional Method 14%
• Scattered Field 14%
• Field Measurement 14%
• Computation Complexity 14%
• Offline Training 14%

Chemical Engineering

• Deep Learning Method 100%
• Neural Network 14%