Table 2 LUT parameters for the deep-CE
From: Adaptive resource optimization for edge inference with goal-oriented communications
\(\rho\) | \(G(\rho ) \,\, [\%]\) | \(\frac{1}{J_{c}}\) \([\frac{C}{DU}]\) | \(\frac{1}{J_{zip}}\) \([\frac{C}{DU}]\) | \(J_{d}\) \([\frac{DU}{C}]\) |
|---|---|---|---|---|
2 | 96.3 | \(4.454 \times 10^{6}\) | \(1.05\times 10^7\) | \(6.66\times 10^{-8}\) |
4 | 97.3 | \(6.46\times 10^6\) | \(5.17\times 10^6\) | \(8.59\times 10^{-8}\) |
8 | 95.1 | \(7.752\times 10^6\) | \(3.38\times 10^6\) | \(8.98\times 10^{-8}\) |
16 | 92.2 | \(8.568\times 10^6\) | \(1.90\times 10^6\) | \(9.47\times 10^{-8}\) |
32 | 88.1 | \(9.486\times 10^6\) | N/A | \(1.05\times 10^{-7}\) |
64 | 82.0 | \(1.02\times 10^7\) | N/A | \(9.8\times 10^{-8}\) |