In rather extensive testing of LTE Cat-M1 with BG96 I have noticed serious faults in data transfer over UDP protocol.
Test setup:
PC with LTE modem and private APN SIM card in that -> fixed IP address of private address space
In PC there is a simple UDP echo server (based in own code) in a continuous listening/echo loop
BG96 with same series private APN SIM card -> fixed IP
In BG96 basic network setup: Cat-M1 network connection established
AT commands to open UDP connection and to send data over it:
AT+QIOPEN=1,2,“UDP”,“10.18.146.24”,4568,7090,1
AT+QISEND=2,10,“10.18.146.24”,4568
Enter 10 characters, for example “1234567890”, and accept by clicking Return.
After this data goes to echo server and is sent back, which causes following notification in BG96:
+QIURC: “recv”,2
Data can be retrieved with the command:
AT+QIRD=2
The problem is that it seems that data is not sent to the network on every attempt. With NB-IoT it works perfectly well but with Cat-M1 less than half of the attempts are successful.
Any ideas what could cause this?
As you said, with NB_iot network , it work fine , so Pls check your network quality of CATM with the below AT cmd, for the detail , pls kindly open <Quectel_BG96_AT_Commands_Manual>
Here are the signal info for both technologies:
+QCSQ: “CAT-M1”,-74,-97,171,-7
+QCSQ: “CAT-NB1”,-82,-91,205,-9
This goes a bit off the topic, but…
For 2G RSSI is the same as for Cat-M1, which makes sense. But how about the difference between Cat-M1 and NB-IoT: how it can be that huge, -74 vs. -82 dB?
This also makes me wonder how much should be counted to those raw dB values in comparison? In other words, are both signals measured in the same way, which would reflect the actual capability to transfer data over the network in the poor signal?
In measurement trips, in various challenging locations, we have noticed that the NB-IoT signal really penetrates through even if 2G or LTE cannot be detected.
On the other hand testing on the sea (boat) for example indicates that the NB-IoT signal is not reaching the device as well with NB-IoT as it does with LTE. Perhaps missing signal-reflections in the open sea could be one reason? Or perhaps interference between multiple base stations?
Cat-M1 testing must be continued to find out the reason why data is occassionally not sent.
