Spatial Transformer Network-Based Automatic Modulation Recognition of Blind Signals

Table 2

Comparisons of the number of parameters and training time for different methods.

Method	Parameter number	Training time (s)
Proposed	4.9 × 10^6	319
CLDNN	5.8 × 10^5	1654
CNN	1.6 × 10^6	297
ResNet	1.8 × 10^6	334