Browse Source

Update README.md

main
Ergun Simsek 2 years ago
committed by GitHub
parent
commit
ffb8b8090a
No known key found for this signature in database GPG Key ID: 4AEE18F83AFDEB23
  1. 24
      README.md

24
README.md

@ -1,16 +1,16 @@
# photodetectors # photodetectors <br>
MUTC Photodetector Dataset MUTC Photodetector Dataset <br>
This dataset lists four performance metrics of 1755 MUTC photodetectors with 17 layers. This dataset lists four performance metrics of 1755 MUTC photodetectors with 17 layers. <br>
The semiconductor types of these layers are fixed and they can be found in https://doi.org/10.1364/OE.27.003717 The semiconductor types of these layers are fixed and they can be found in https://doi.org/10.1364/OE.27.003717 <br>
**Columns** **Columns**
phasenoise : true phasenoise of the photodetector (Unit: dBc/Hz) phasenoise : true phasenoise of the photodetector (Unit: dBc/Hz) <br>
phasenoise_15mA : hypothetical phasenoise of the photodetector if the output current was equal to 15 mA (Unit: dBc/Hz) phasenoise_15mA : hypothetical phasenoise of the photodetector if the output current was equal to 15 mA (Unit: dBc/Hz) <br>
current : average current of the photodetector (Unit: mA) current : average current of the photodetector (Unit: mA) <br>
IR_max : maximum value of the impulse response of the photodetector IR_max : maximum value of the impulse response of the photodetector <br>
decay_time: the time it takes for the impulse response to decay from its maximum value to 1% of that maximum value (Unit: ns) decay_time: the time it takes for the impulse response to decay from its maximum value to 1% of that maximum value (Unit: ns) <br>
t1 ... t17 : thickness each layer (Unit: nm) t1 ... t17 : thickness each layer (Unit: nm) <br>
d1 ... d17 : doping concentration of each layer (Unit: cm^-3) d1 ... d17 : doping concentration of each layer (Unit: cm^-3) <br>
**Forward Problem:** Predicting Performance Metric From Design Parameters **Forward Problem:** Predicting Performance Metric From Design Parameters <br>
The four machine learning algorithms, linear regression, k-nearest neighbor, random forest, and an artificial neural network, are used to predict performance metrics (phasenoise, current, IR_Max, and decay_time) from the design parameters (thicknesses and doping concentrations) The four machine learning algorithms, linear regression, k-nearest neighbor, random forest, and an artificial neural network, are used to predict performance metrics (phasenoise, current, IR_Max, and decay_time) from the design parameters (thicknesses and doping concentrations)

Loading…
Cancel
Save