Table 2 Simulation results of case B on different subsets with ResNet block depth of 5, 10, 15.
From: Super resolution DOA estimation based on deep neural network
Block depth | \(\hbox {N} = 5\) | \(\hbox {N} = 10\) | \(\hbox {N} = 15\) | ||||||
|---|---|---|---|---|---|---|---|---|---|
Signal number | P | R | F | P | R | F | P | R | F |
1 | 0.9985 | 0.9959 | 0.9972 | 0.9991 | 0.9987 | 0.9989 | 0.9996 | 0.9994 | 0.9995 |
3 | 0.9894 | 0.9884 | 0.9889 | 0.9944 | 0.9934 | 0.9939 | 0.9958 | 0.9956 | 0.9957 |
5 | 0.9259 | 0.9203 | 0.9231 | 0.9350 | 0.9300 | 0.9325 | 0.9409 | 0.9406 | 0.9408 |
7 | 0.8131 | 0.7788 | 0.7956 | 0.8289 | 0.8013 | 0.8149 | 0.8379 | 0.8226 | 0.8302 |
