• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 16 subscribers
  • Views 898 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Frequency separation versus Baudrate

Enrico S.
Intellectual 290 points
Other Parts Discussed in Thread: CC1020

I'm designing a system with the CC1020 using SmartRF studio (thanks your to previous help, also).

I'm trying to understand how what happens if the frequency separation of a 2-FSK system is lower than the baudrate.

I found that this frequency should be, for narrow band systems, ideally the same as the baudrate. Could be greater than this, and I'm obtaining a wide band system. That's ok.

My question is what happens if the separation is lower than the baudrate? I'm asking this because the CC1020 transceiver, with high baudrate (153600), with a frequency separation of more than 153kHz, will not obtain a proper configuration with its SmartRF studio configuration. The option is to reduce the crystal uncertainty or the frequency separation, while the channel spacing is already set to maximum (500kHz). 

My system is half duplex, with same frequency in RX and TX (work alternatively on the same chip). At this moment, if too much wideband, could be still ok.

Thanks to all.

  • 0
    TI__Mastermind 39860 points
    Theoretically, there is an optimum separation/datarate setting if you simultaneously minimize the receiver filter bandwidth. Every halving of receiver filter bandwidth increases sensitivity with 3 dB whereas sensitivity vs separation/datarate decreases with about 1.5-2.5 dB per halving down to a certain limit where the loss increases very fast. In our experience a modulation index (=separation/datarate) = 1 is a good design compromise.

    Using a mod index of less than 1 is not a problem (153 kHz separation and 153.6 kbps data rate)
  • 0 in reply to Sverre
    Intellectual 290 points

    Sverre said:
    sensitivity vs separation/datarate decreases with about 1.5-2.5 dB per halving down to a certain limit where the loss increases very fast

    Firstly, thanks for the answer.

    But I don't understand (grammatically) the quoted sentence: "separation/datarate" is a ratio? And you mean that a plot of sensitivity versus this ration decrease of 1 or 2.5 dB per halving the separation/datarate ratio?

    If what I stated is correct, at 153.6kBaud at 76.8kHz of separation, the sensitivty is lowered by 2dB? And if so, the loss "increased very fast" is related to what and how can be quantified?

    EDIT: in section 12.4 of datasheet there is what we are talking about. My question was to understand also if it was possible to achieve better performances at these rates. And here why the need to understand better these relations.


    Thanks.

  • 0 in reply to Enrico S.
    TI__Mastermind 39860 points
    Assume the RX filter BW is fixed. The data rate is 100 kBaud
    With 50 kHz deviation (= 100 kHz separation) the mod index = 1. The sensitivity is -Y dBm
    With 25 kHz deviation (= 50 kHz separation) the mod index = 0.5. The sensitivity is (-Y + 1.5/2.5) dBm

    If you lower the modulation index (i.e. separation) much below 0.5 the sensitivity you will take a hit on sensitivity.

    To find the optimum separation/data rate setting vs RX filter BW for your application you also need to take into account the crystal tolerance and do some real measurements. I suggest you use the settings you get from SmartRF Studio if you are to operate at 153.6 kBaud.
  • 0 in reply to Sverre
    Intellectual 290 points

    Ok, thanks. On this topic I have only two questions left, if possible:


    1. The most important question: I will consider a doppler on the carrier of about 10kHz, should this taken into account in the channel spacing and/or filter BW?

    2. In table 19 of datasheet, using 2.4kBaud with deviation higher than 9.9kHz (with the appropriate BW ecc), will bring sinsitivity lower than -118dBm?  Or these are side conditions which should not be taken into account?

  • 0 in reply to Enrico S.
    TI__Mastermind 39860 points

    1. A shift in the carrier frequency must be taken into account. The following post discusses RX filter BW vs crystal tolerance:  

    2. If you increase the deviation @2.4 kBaud to +/-9.9 kHz you need to increase the RX filter BW to approx 28 kHz

    Increasing the RX filter BW from 9.6 kHz to 28 kHz gives 10*log(28/9.6) = 5 dB degradation in sensitivity (everything else being the same)

    Increasing the deviation from 2 kHz to 9.9 kHz will improve the sensitivity for the same RX filter BW. I have not measured this, but I will expect close to -118 dBm sensitivity also @2.4 kBaud, +/-9.9 kHz, 28 kHz RX filter BW. Furthermore, increasing the RX filter BW will degrade the close-in selectivity.