Skip to main content

IBF Tests

All IBF tests should be run with the E2E controller.

P2P

Note: Auto PBF after IBF available in M37+. Heat map data is based on IBF.

PUMA_RF_IBF-P2P-1 DN-DN Assocication

Description: P2P heat map test (DN to DN) on long link.

Procedure:

  1. Enable BF stats using r2d2 command on both nodes.
  2. Associate DN-I ↔ DN-R.
  3. Record the phyPeriodic.txBeam used each time.
  4. Delay of 30 seconds and send 50 (TBR) pings once a second (TBR).
  5. Disassociate DN-I ↔ DN-R link.
  6. Repeat 50 times.
  7. Measure phystatus.srssi and LQM in each iteration.

Passing:

  1. No failed assocs, pings, disassocs.
  2. Figure out the max and verify that beam indices pick the max LQM.
  3. Note: 5-6 sec difference is acceptable verifying the beam indices and the max LQM.

PUMA_RF_IBF-P2P-2 DN-DN Assocication

Description: P2P heat map test (DN to DN) on short link.

Procedure:

  1. Enable BF stats using r2d2 command on both nodes.
  2. Associate DN-I ↔ DN-R.
  3. Record the phyPeriodic.txBeam used each time.
  4. Delay of 30 seconds and send 50 (TBR) pings once a second (TBR)
  5. Disassociate DN-I ↔ DN-R link.

Passing:

  1. No failed assocs, pings, disassocs.
  2. Same beams must be selected every time.

Description: Ignite P2P link in coffin setup multiple times and verify link comes up with same beams.

Procedure:

  1. Enable BF stats on both nodes.
  2. Set xdB attenuation on the link (where x is enough to not bring the link up).
  3. Ignite P2P link in coffin.
  4. Reset attenuation back to 0dB.
  5. Delay of 30 seconds and send 50 (TBR) pings once a second (TBR).
  6. Repeat 10 times.
  7. Plot the phystatus.srssi in each iteration.
  8. Measure LQM in each iteration.

Passing:

  1. Verify that the E2E controller attempts to ignite the P2P link when attenuation is high till the command times out.
  2. Verify LQM diff is no more than +/-1 between each iteration.

Description: Ignite P2P link in coffin setup with high attenuation and verify ignition succeeds with 0dB attenuation.

Procedure:

  1. Enable BF stats on both nodes.
  2. Set xdB attenuation on the link (where x is enough to not bring the link up).
  3. Ignite P2P link in coffin.
  4. Reset attenuation back to 0dB.
  5. Delay of 30 seconds and send 50 (TBR) pings once a second (TBR).
  6. Repeat 5 times.
  7. Plot the phystatus.srssi in each iteration.
  8. Measure LQM in each iteration.

Passing:

  1. Verify that the E2E controller attempts to ignite the P2P link when attenuation is high till the command times out.
  2. Verify LQM diff is no more than +/-1 between each iteration.

Description: Ignite P2P link in coffin setup with 0dB attenuation and verify ignition fails with high attenuation.

Procedure:

  1. Enable BF stats on both nodes.
  2. Set 0dB attenuation on the link.
  3. Ignite P2P link in coffin.
  4. Delay of 30 seconds and send 50 (TBR) pings once a second (TBR).
  5. Set attenuation to high value.
  6. Repeat 5 times.
  7. Plot the phystatus.srssi in each iteration.
  8. Measure LQM in each iteration.

Passing:

  1. Verify that the E2E controller attempts to ignite the P2P link when attenuation is high till the command times out.
  2. Verify LQM diff is no more than +/-1 between each iteration.

PUMA_RF_IBF-P2P-6 Off-boresight beam selection

Description: Ignite a P2P link that is off-boresight and check beam.

Procedure:

  1. Enable BF stats on both nodes.
  2. Associate the P2P link.
  3. Delay of 30 seconds and send 50 (TBR) pings once a second (TBR).
  4. Repeat 5 times.
  5. Plot the phystatus.srssi in each iteration.
  6. Measure LQM in each iteration.

Passing:

  1. Verify that the beam idx picked each time correlates with the actual angle between the nodes.
  2. Verify LQM diff is no more than +/-1 between each iteration.