Changes between Version 120 and Version 121 of Other/Summer/2025/mlCoexist


Ignore:
Timestamp:
Jul 28, 2025, 6:54:01 PM (13 days ago)
Author:
aw1086
Comment:

—

Legend:

Unmodified
Added
Removed
Modified

  • Other/Summer/2025/mlCoexist

    v120 v121  
    161161
    162162- MATLAB code modification:
    163  - Added a variability margin to the signal gains to simulate the uncertainty in physical transmissions and obtain multiple samples from a single scenario
    164  - Edited the y-axis (frequency range: 1400 MHz–1427 MHz) to illustrate only the section of signals that leaked into the L-band, in order to prevent the CNN model from picking up insignificant details during training (L: original; R: cropped)
     163 1. Added a variability margin to the signal gains to simulate the uncertainty in physical transmissions and obtain multiple samples from a single scenario
     164 2. Edited the y-axis (frequency range: 1400 MHz–1427 MHz) to illustrate only the section of signals that leaked into the L-band, in order to prevent the CNN model from picking up insignificant details during training (L: original; R: cropped)
    165165[[Image(https://www.orbit-lab.org/raw-attachment/wiki/Other/Summer/2025/mlCoexist/original.png, 32%)]]
    166166[[Image(https://www.orbit-lab.org/raw-attachment/wiki/Other/Summer/2025/mlCoexist/cropped.png, 32%)]]
    … …  
    172172
    173173**Progress:**
    174 [[Image(https://www.orbit-lab.org/raw-attachment/wiki/Other/Summer/2025/mlCoexist/newDataTable.png, 56%)]]
     174- Created new numpy data table for regression model training (3 columns: RFI Scenario | Transmitting Signal Spectrogram | Brightness Temperature)
     175 - Since brightness temperatures of all samples from a single RFI scenario (band, central frequency, resource block, gain) are uniformly distributed, we ordered the Tb values obtained from the reference testbed L1B data from lowest to highest, and matched them with the 10 spectrograms generated with different marginalized signal gain values (uniform from -0.276 to +0.317) that simulate variation in physical transmission
     176 - With this method, we are able to obtain more data points to train and test the CNN, instead of using the exact gain values (-20, -10, 0, 10, 20) to form only one sample per scenario, leaving SN2 to SN10 from the L1B data unused
     177[[Image(https://www.orbit-lab.org/raw-attachment/wiki/Other/Summer/2025/mlCoexist/newDataTable.png, 52%)]]
    175178\\
    176179\\