• State TI Thinks Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 2 answers
  • Subscribers 54 subscribers
  • Views 291 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

LAUNCHXL-CC1310: Strategy for working in noisy environment

LAUNCHXL-CC1310: Noisy 915 Environment.

Oliver Lewis
Intellectual 635 points
Part Number: LAUNCHXL-CC1310
Other Parts Discussed in Thread: CC1310

Hi,

This is a continuation of a thread posted a few months ago (https://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1112545/launchxl-cc1310-strategy-for-working-in-noisy-environment). Firstly, thank you for your previous responses. While they led us to perform further tests, we haven't managed to solve this issue. 

From the last post, I have a question regarding blockers, which I didn't pick up on last time. The comment from engiNerd read:

"By the way, if you have a very strong blocker, it will desensitize the whole radio such that it will look like you have a flat noise floor. It would be interesting to check the noise floor with a narrowband setting (like 10kHz) - it is still flat?"

Firstly, I'm not entirely sure what is meant by "blockers" in the given context. I did a little search for blockers and found a quote on your tech summary for the cc1310. The summary reads:

"Long range operation, the sensitivity parameter is -110dBm at data rate of 50kbps and down to -124dBm when the data rate is 0.625kbps. Interference from other wireless communications can be overcome with 90dB of blocking." 

Does this quote mean that a 90dB blocking needs to be added? If so how?

In any case, we are still seeing interference/noise across the whole 915 band:

  • Can't achieve a sensitivity of lower than -80dBm (~range of <150m).
  • The radio profile we use is WB-DSSS/FEC/500kbps/D=8/K=4/30ksps/RX BW 622KHz - taken straight from SmartRF Studio without changes.
  • Using launchpads instead of our h/w shows the same noise floor on every channel at 915 as our hardware and has the same sensitivity problems.
  • Sites we have tested are across the whole of Colorado and at least one site in Texas. All test sites are remote Farms in arid conditions. 
  • When we test the same kit here in the UK we get 800m and a sensitivity down to -107dBm (exact kit was shipped to the US to rule out differences in h/w).
  • Channel hopping doesn't help.
  • CCA before sending doesn't help.
  • We do see improvements if we narrow the bandwidth to sub ~80KHz, but that's not legal to use in the US.

Our current suspicions are:

  • We do not believe there is other radio interference across the whole permissible bandwidth and across the whole of Colorado and into Texas - that's really unlikely.
  • The environment may be causing issues where we are testing, maybe the arid conditions or the rock/stone ground. This could maybe cause excessive reflections or high static condition (large amount of dust particles carried in wind) - we're not sure how to test that though.
  • The front end of the RX side is configured too sensitively (e.g. loop bandwidth lock needs to be higher, AGC reference level is out, front-end filter is not filtering sufficient).

Is there anything further you might be able to suggest to help us investigate and solve this issue?

  • What's a 90dB blocker (see above)?
  • Any tweaks to the overrides or set up parameters what would help in a noise environment?
  • Anything? Anything at all?

Sincerely,

- Oliver

  • 0
    TI__Mastermind 30890 points

    Hi Oliver,

    If there is a general RF noise across the whole band at -80 dBm then you can double check this with a Spectrum Analyzer (SA) and an antenna connected.

    Do you have access to a SA and an antenna so you can measure the noise floor at 915 MHz ?

    For the frequency span on the SA, use the same as the Rx bandwidth frequency of the CC radio. 

    What is the noise floor you are seeing on the SA ?

    Regards,

       Richard

  • 0 in reply to RGW
    Intellectual 635 points

    Hi Richard,

    Thanks for your reply, hope my queries aren't interrupting any thanksgiving celebrations you may have planned.

    We see a noise floor of approx -80dB across the whole spectrum and  approximately the same for a span ranging 600KHz to 4MHz. This is pretty consistent across at every site we've been to - we even tried going to remote a desert. In populated areas, we can see the occasional LoRaWAN or similar network, but this is not present or obvious at our installation sites. The launchpad reports the same noise floor if we sit on any frequency within that band.

    We see the same noise floor here in the UK where we do seem to get a great range.

    What would you suggest, given that DSSS should happily exist well below/within the noise floor?

    In my previous exchanges, one of your engineers suggested looking at LRM mode. I haven't done this yet (the baud rate is a little low for us), but could you confirm that LRM even legal to use under FCC - your examples only us a 10KHz overall bandwidth?

    Regards

    - Oliver

  • 0 in reply to Oliver Lewis
    TI__Mastermind 30890 points

    Hi Oliver,

    How are you measuring noise floor ?

    I would check with an SA to confirm that the noise floor and make sure the span is set to the same as the Rx bandwidth. If the noise floor is as high as -80 dBm you will still see this in the lab.

    I do not believe the noise floor is -80 dBm across the complete "915" MHz band otherwise this would affect all products working in this ISM band with poor range.

    Regards,

       Richard

  • 0 in reply to RGW
    Intellectual 635 points

    Hi Richard,

    We normally measure the noise floor with an SA set to a 622KHz span, centered on any one the channels we use. We have also used the continuous RX on the launchpads to do the same.

    Just doing a noise floor test here in my office (UK) it's actually fine -110dBm (apart from the off mobile signal). I was under the impression it was similar to the US, alas I stand corrected.

    I am going to instruct our OPs person to re-do the SA scan across the whole 915 band again on Monday and I shall do the same here. 

    Regards,

    - Oliver

  • 0 in reply to Oliver Lewis
    Intellectual 635 points

    Hi Richard,

    I have had the field ops guys in the US redo an SA scan at one of the sites and the noise floor there is indeed around -80dBM (same settings on SA as mine). This appears across the whole 902-928MHz band. Interestingly, there seems to be very little attenuation inside a house and reducing the span also shows no change in the noise floor (tmeasured from 622MHz to 10KHz).

    This mimics the same floor we were seeing on our remote target farms and even in the dessert. In any case, this noise floor is killing our radio. There are no notable signs of other transmissions either - e.g. the SA will happily pick up LoRaWAN - it's just a flat floor. While I find this equally implausible across the whole of Colorado and some of Texas, out reception does cut out when we hit the noise floor, where we are a good 20dBm below the sensitivity we would be expecting.

    Regards,

    - Oliver

  • 0 in reply to Oliver Lewis
    TI__Mastermind 30890 points

    Hi,

    I'm asking colleagues in Dallas to measure the noise floor just to check. 

  • 0 in reply to RGW
    TI__Expert 3970 points

    Hi Richard/Oliver,

    The results were measured on the cc1310 launchpad.

    See picture below for results from our RF chamber. The results we're measuring is ~100dBm.

    See below for results from outside the RF chamber. The results are similar.

  • 0 in reply to Bun Kuch
    Intellectual 635 points

    That's interesting, thanks for doing the scan, much appreciated.

    That is almost identical to what we see here in the UK, but not at any of our sites in Colorado.

    Is there anything you might recommend at this point? 

    Sincerely,

    - Oliver

  • 0 in reply to Oliver Lewis
    TI__Expert 3970 points

    Hi Oliver,

    Were you able to perform a conducted test through an SMA or U.FL connector in the same arid environment to rule out hardware issues? It would be good to have a reference data.

    Best regards,

    Bun

  • 0 in reply to Bun Kuch
    Intellectual 635 points

    Hi Bun,

    Thanks for the suggestion. Our field engineer in the US is currently in the UK, so our ability to try this is temporarily delayed (until ~12th). That's an easy enough test to do though and I will get onto it asap.

    What would this show us, that a comparison to the launchpads doesn't? Remember, we see the same poor performance and read the same noise floor in both versions of the cc1310 evks too - SmartRf6 EBK with plug-in cc1310 module, and the US version of red launchpad variant with L-shaped antenna.

    Are you think that there could be some inherent environmental susceptibility in the RF front-end?

    Thanks, Bun. You'll need to give me about a week to organise this.

    Sincerely,

    -Oliver