Skip to main content
EURASIP Journal on Wireless Communications and Networking

Table 2 Layers of SigZSLNet and the output dimensions of each layer

From: An automatic modulation classification network for IoT terminal spectrum monitoring under zero-sample situations

Module

Layer

Output

CNN

Input Layer

1024 × 2

Conv1D (filters 64, size 1 × 8)

1024 × 64

Maxpooling (size 1 × 2)

512 × 64

Conv1D (filters 64, size 1 × 8)

512 × 64

Maxpooling (size 1 × 2)

256 × 64

Conv1D (filters 64, size 1 × 8)

256 × 64

Maxpooling (size 1 × 2)

128 × 64

Conv1D (filters 64, size 1 × 8)

128 × 64

Maxpooling (size 1 × 2)

64 × 64

Conv1D (filters 64, size 1 × 8)

64 × 64

Maxpooling (size 1 × 2)

32 × 64

Flatten

2048

Fully connected layer

128

G

Input layer

64

Fully connected layer

256

Leaky_relu

–

Fully connected layer

128

D

Input layer

64

Fully connected layer

256

Leaky_relu

–

Fully connected layer

128

Classifier

Input layer

128

Fully connected layer

128

PReLU

–

SoftMax

7

  1. G is the generator module, and D is the discriminator module
Back to article page