Research Article
Multisource DOA Estimation in Impulsive Noise Environments Using Convolutional Neural Networks
Table 4
DOA1 subnetwork architecture.
| Layer and activation | Output shape | Parameter |
| L0: | Input | 8×8 | 0 | L1: | Convolution + Relu | 32 × 8 × 8 | 320 | L2: | Convolution + Relu | 32 × 8 × 8 | 9248 | L3: | Convolution + Relu | 16 × 8 × 8 | 4624 | L4: | Convolution + Relu | 16 × 8 × 8 | 2320 | L5: | Flatten | 1024 | 0 | L6: | Dense + Relu | 8 | 820 | L7: | Dense + Relu | 1 | 9 |
|
|