Hello, I’m Santosh. I’m new to LoRa P2P mesh networking and am currently working on a project utilizing RAK3172 modules.
In my LoRa P2P mesh network using RAK3172 modules, I have a node transmitting 44-byte payloads (including a 2-byte CRC). The other five nodes send 20-byte payloads (also including a 2-byte CRC).
All nodes are configured with the following parameters:
- Spreading Factor (SF):10
- Bandwidth (BW):125 kHz
- Frequency:865 MHz
- Coding Rate (CR): 4/5
- Preamble Length: 8
- TX Power: 22 dBm
- Transmission Delay: Random between 0–3000 ms
The issue arises when the node sends the 44-byte payload: the receiving nodes have bit flips specifically in the 2-byte CRC portion of the payload. This problem doesn’t occur with the 20-byte payloads from the other nodes.This is unexpected, given that the theoretical maximum payload size at SF10 is 51 bytes.
Additional Observations: Notably, this CRC bit-flipping issue only manifests when the node operates within the full mesh network. When the device is isolated or when a few other nodes are turned off, the issue does not occur.
Troubleshooting Steps Taken:
- Verified that all nodes share identical configuration parameters..
- Ensured firmware versions are consistent across all nodes.
- Confirmed that the issue is isolated to the 44-byte payload transmissions.
- Experimented by adjusting the coding rate specifically for the node in question, testing all available coding rates—CR 4/5 (0), 4/6 (1), 4/7 (2), and 4/8 (3)—but these adjustments did not mitigate the problem.
Questions:
- Is there a recommended maximum payload size for reliable transmission at SF10, considering the current configuration?
- Are there known issues when using different payload sizes across nodes in a LoRa P2P network?
- Could the issue be influenced by network congestion or collisions inherent in the mesh topology?
- Are there any best practices to prevent such CRC bit-flipping issues in similar scenarios?
I would greatly appreciate any guidance or suggestions on how to resolve this CRC bit-flipping issue in the 44-byte payloads. If anyone has encountered similar problems or has insights into potential causes and solutions, your expertise would be invaluable.
Thank you in advance for your support!