Hi
I am wondering if you guys have a way of measuring the SINR (or SNR) using the RFM22. I have been experimenting with the device for some time now and so far and it is
giving some weird behavior, I hope we guys can shed the light on what
is going on.
I did the following two setups:
1- RFM22 as a spectrum scanner:
In this setup, I run the code that is provided in the library and it works as expected i.e., as I increased the level of interference on the target frequency, the RSSI goes up, regardless of the shape of the interfering signal as long it is on the target frequency. This is why I think the RFM does energy detection when it does not find a valid preamble (i.e., it measures the energy of EVERYTHING that is on that frequency).
2- RFM22 and the RSSI of a successfully received packet:
In this setup, I had two Tx-Rx pairs on the same frequency with matching parameters and I had a jammer which I turned on and off to see how would that affect the RSSI.
This is were things went a lil weird. Before sending a valid packet (from the Tx to the Rx), upon calling the readRssi(), I was getting the usual behavior (more interference by the jammer --> higher RSSI). However when I send a packet and it is received successfully, I am calling the lastRssi() which reports the RSSI of the packet that was lastly received correctly (according to the library it is computed during the preamble).
I noted that when I increase the power of the interfering signal, the measured RSSI by lastRssi() goes down.
So how the chip is doing these two behaviors? one thing that comes to me is that it is using two different approaches to calculate the RSSI (i.e., one for no valid reception and the other for valid reception). The other question is that, what is the valued that is reported by lastRssi() (we know it is the RSSI of the last correctly received packet) but how was it calculated and what does it represent is it the SINR, SNR or it is just the power of the signal that is being received?
In other-words, does an Rssi that is reported by "lastRssi()" represent the true energy of the signal (without the energy of the noise) or what?
Thx again for the sweet library.
Mike.