Research Article

Semi-supervised Learning for Automatic Modulation Recognition Using Haar Time–Frequency Mask and Positional–Spatial Attention

Table 5

The recognition accuracy of HTF-PSA-SSL under different label rate: 1% (88), 5% (440) and 10% (880).

SNR (dB)1% (88)5% (440)10% (880)

−189.09%9.55%9.73%
−1220.00%20.36%20.73%
−653.09%54.73%55.36%
087.64%89.82%89.73%
692.09%92.64%93.09%
1292.09%93.00%93.55%
1892.36%93.82%94.09%