Hi there,
I'm performing an throughput analysis using an nRF52 development board (BMD-350) and a Raspberry Pi 4B (Ubuntu 20.04, BlueZ 5.54) with an BLE 5.0 adapter.
For measuring the throughput of the connection I'm using 10 attributes with 247 bytes (ATT_MTU-4) each, with notifications enabled.
I adjust the frequency of updating the attributes and log the maximum throughput measured on the Raspberry Pi.
I've tried modifying multiple parameters on the connection, but despite this, the maximum throughput I was able to achieve was 200kbps (for LE 2M, but the result was the same using LE 1M).
Oddly, the throughput seems to only depend on the connection interval (albeit very slightly) and with the MTU size. If I try changing the PHY to LE 2M, I don't observe any change in the throughput, which leads me to believe there is another unidentified bottleneck on the connection which I just can't seem to find.
These are the connection parameters which gave me the best throughput:
- Connection Interval: 7.5 ms;
- Slave Latency: 0 ms;
- Supervision Timeout: 2000ms;
- MTU: 251 bytes (using Data Length Extension);
I'm new to nRF development so any help is appreciated, thank you in advance!
