I am new to LimeSDR platform. I am trying to enable 802.15.4e TSCH protocol using GNU Radio with LimeSDR-USB A. Because the protocol is time slotted protocol I have strict timing constraints. I am trying to do a round trip delay analysis of the test setup using RF loopback enabled on the FPGA. I have noticed the following thing (correct me if I am wrong):
The IQ data stream I am sending never comes back as it is, the received stream has reduced amplitude and also the digital sample values are changed. Because I am not using the LMS7002M chip shouldn’t the RF loopback enable me to receive the same data stream?
Also, I tried using the measureDelay.py to measure the delay but it fails because the argMax index and corrIndex are not converging. I have enabled the loopback by writing 0x55 to the FPGA register 0x0017 ( I took the values from the Limesuite Implementation) but the results are still not converging.
Anyway, RF loopback test means, that you have this chain:
PC --> FPGA --> LMS7002M Tx (ADC, …, RF) --> LMS7002M Rx (RF, …, DAC) -> FPGA -> PC
As you can see, RF is included in the chain. Hence no wonder that your signal gain will be different and received samples doesn’t match with those were sent.
Message Source -> GNU Radio -> FPGA TX----> FPGA RX -> GNU Radio -> Message Sink.
The gnu radio is running 802.15.4 MAC and PHY layers. So when I send the message I measure the system time lets say t1 and when I receive my message I get the time t2. So I measure the overall round-trip latency as t2-t1.
Next, I want to measure the timing from USBMon logs, that’s why I was curious to what the " RF loopback CH A" and " RF loopback CH B" as found in Board Controls of Limesuite actually means as it would help me figure out a method to identify the packets correctly when I receive them with USBmon.
Also is there any way to disable the LMS7002M chip and loop the stream back from the FPGA?
Thanks a lot for the help. I will try it out. Just to clarify these addresses are LMS7002M registers and using these settings I won’t need to set the 0x0017 address using the Board SPI, right?