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.

CC1120 Sensitivity Calculations

Other Parts Discussed in Thread: CC1120, CC1190, CC1125

For the CC1120, Is there a calculation tool to determine receiver sensitivity at bit rates and RX bandwidths other than those listed in the data sheet? Perhaps a calculation tool that helps determine optimum RX bandwidth TX deviation etc… based on desired bit rate?

  •  

    Hi,

    You can use the following formula

    Sensitivity = -174+NF+10*Log(BW)+S/N

    Thanks,

    PM

  • For applications that allow changing the RX filter bandwidth, there is theoretically an optimum deviation/datarate setting if you simultaneously minimize the receiver filter bandwidth. Every halving of receiver filter bandwidth increases sensitivity with 3 dB whereas sensitivity vs deviation/datarate decreases with about 1.5-2.5 dB per halving down to a certain limit where the loss increases very fast. When we optimized CCxxxx settings, we select the frequency deviation that gives the best sensitivity (i.e smallest RX filter BW) and also allows for a certain crystal frequency offset. We always recommend to, if possible use the preffered settings from SmartRF Studio, at least use them as a starting point before starting to alter parameters. See also design note DN005 that explaines for the CC11xx products how the sensitivity changes with frequency accuracy, frequency offset, filter bandwidth etc.

  • Great,

    What S/N would you put in the formula for a 1% BER? Do you happen to have the NF for the CC1120EM? I used 3db for NF and 12db for S/N and it seems fairly close to the known specs from the datasheet. I'd like to dial it in a little closer if I can.

    Thanks,

    Eric

  • Hi,

    For the CC112x would use NF of 8 dB.

    Regards,

       Richard 

  • The SNR number is dependent on the modulation format. What specifically are you looking for?

  • I'm building a variable rate 3-24Kbps FCC part 15.247 transceiver. RX sensitivity at Rates of 3,6,10 and 20Kbps are of particular interest. Range in a mobile environment is of utmost importance. My understanding is that 2-FSK, 2-GFSK, 4-FSK, 4-GFSK are all options and that FSK rather than GFSK would provide the most range at these rates. I'm open to using a high stability TCXO to keep RX BW as narrow as possible and anything else that will maximize sensitivity. Also planning on using CC1190 or Skyworks front end/PA.

    Any suggestions for maximizing RX sensitivity and what sensitivity I might expect at these rates would be of great help.

  • Yes, link budget can be improved by using CC1190.

    You may use CC1125 for narrowband application. Please refer to our data sheets for Sensitivity figures.

     

    Thanks,

    PM

  • Hi,

    Thanks.That's a good point. No reason to use a 2ppm TCXO when the CC1120 RX filter only goes down to 8K. I have considered the CC1125 however without more bit rate specific RX calculations I cannot make an accurate determination of advantages for my project (I realize the CC1125 has better blocking and selectivity). This is for a 902-928Mhz FHSS system and I have data sheets for CC1120 and CC1125 however they do not include "best case sensitivity" nor optimum deviation and RX bandwidth for  "narrow band" FSK and GFSK at or near the rates I mentioned in the previous post. In order to determine this It seems I would need to know the SNR number for 2-FSK, 4-FSK, 2GFSK and 4-GFSK as well as NF to use for the CC1125. Since it sounds like TI does not have a calculation tool readily available for this. I plan to make a spreadsheet  to compare and optimize with my specific throughput requirements. Then field test with the CC1120DK. So... Can I get the SNR figures to use for all modulation types and the CC1125 NF? Any additional optimization information would be great as well.

    Thanks

    Eric

  • To give you an idea on the sensitivity I did some quick measurements on 2 samples. I have not run PER vs level so it could be that AGC_REF has to be adjusted some. 

    FSK gives better sensitivity but on the other side the spectrum is narrower with GFSK which enables you to use a smaller RX BW. You also have to calculate how wide you can set the RX BW compared to the signal bandwidth given the crystal tolerance. 

    5241.E2E_settings.xlsx