Table 3 PSTNN structure design parameter
From: Spatial and temporal learning representation for end-to-end recording device identification
SFENN | TFENN | |||
|---|---|---|---|---|
DNN | CNN | ResNet | LSTM | Bi-LSTM |
Input (2496) | Input (64*39*1) | Input | Input | Input |
FC (1024) | Conv (6*5*5) | BN+Relu | LSTM(39) | Bi-LSTM(39) |
FC (1024) | pooling | Conv(16*3*3) | LSTM(78) | Bi-LSTM(78) |
FC (1024) | Conv (16*5*5) | BN+Relu | FC(1024) | FC(1024) |
| Â | pooling | Conv(16*3*3) | Â | Â |
| Â | Conv (40*5*5) | BN+Relu | Â | Â |
| Â | pooling | Conv(16*3*3) | Â | Â |
| Â | FC (1024) | BN+Relu | Â | Â |
| Â | Â | Conv(16*3*3) | Â | Â |
| Â | Â | pooling (Input) | Â | Â |
| Â | Â | Add() | Â | Â |