Research Article
CSL-SFNet for Cooperative Spectrum Sensing in Cognitive Satellite Network with GEO and LEO Satellites
Table 1
Description of the system model parameters.
| Parameter | Symbol | Value |
| Training set size (per SNR) | — | 1,000 | Test set size (per SNR) | — | 100 | Batch size | — | 512 | Dropout rate | — | 0.1 | Learning rate | — | 0.0001 | Optimizer | — | Adam | Modulation scheme | — | 16QAM | SNR range | — | −20 to 0 dB | Sample length | | 512 | Transmit power of GES | | 40 dBm | Noise temperature of SU | | 175K | Carrier frequency | | 29.9 GHz | Bandwidth | | 24 MHz | Atmospheric absorption factor | | 0.75 dB | Cloud attenuation factor | | 1.25 dB | Antenna diameter of SU | | 0.3 m | Antenna diameter of GES | | 0.3 m | Antenna efficiency | | 55% | Probability of detection/false alarm | | — | Optimal model parameter of CSLNet | | — | Optimal model parameter of Soft-FusionNet | | — | Maximum gain of the GES’s transmit antenna | | — | Maximum gain of the SU’s receive antenna | | — | Noise uncertaint factor | | — |
|
|