...

1. David Burnett

2. Fil Maksimovic

3. Austin Patel

4. Tarak Arbi

5. Kristofer PISTER

6. Alex Moreno

Action Items

Notes

• Fil Maksimovic SCuM can stay synched at 10 s intervals with 1.5 ms guard time.

  • Network was OpenMote (root) → scum → scum. Adding 1 more scum is next.

  • Maybe getting an OpenMote to join the network with a SCuM parent after a chain of 2-3 scums.

  • Would require using opentimer instead of direct access to capture/compare registers

  • Beware low PDR (60-70%) if the receiver is on for a long time vs. being duty cycled. Root cause as yet unknown.

    • Also helpful is to powercycle the radio every ~100 ms.

    • Open question is whether the missed packets are periodic, e.g., 3 OK 1 missed etc.

  • Austin Patel interested in a star network, OpenMote (root) → 5x scum, current results should support that network.

    • Data rate will be limited because OpenMote can only Rx from one scum at a time.

    • If OpenWSN is using 10 ms slots, mean latency should be 25 ms. Packet loss must be accounted for too.

    • Music needs 1 ms latency, typical gaming is 20-70 ms response, high performance gaming is 1 ms. Typical human eye to hand reaction time is 70 ms.

• Tarak Arbi presented slides about BPSK demodulation

  • Per Fil, looks like bank 1 GPOs 8-15 output have I_LC<1:4> and Q_LC<1:4> I think those are the two 4bit ADCs. We may be able to use those GPOs to get sampled IF output and try to do BPSK demod.

• Alex Moreno slides on freq vs vbat relationship (supply pushing)

• SSCS article on wireless BMS https://ieeexplore.ieee.org/document/9805643