• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 55 subscribers
  • Views 687 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.

CC1120: BER Testing with SmartRF Studio

Zebryk
Expert 1240 points
Part Number: CC1120

Does anyone know of a way to use SmartRF Studio for BER Testing?

NOT for PER measurement which usually is preferred, but for strictly examination data streams having no preamble, sync or other content?

I would like to inject either a Random or 0aAA/0x55 stream of bits from a TX "window" and observe the reaction on the RX "window".

This would enable the study of various combinations of TX-RX Hz Offset, BPS, Deviation, TXbw, RXbw, etc..

Thanks,

--

Jay Zebryk

  • 0
    TI__Guru**** 317180 points
    CC1120 is a packet based radio so testing without preamble/ sync will not give you the optimal result. Transparent mode is an option but then you can't use SmartRF Studio to read out the results and the dataoutput will have some jitter meaning that the receiver MCU should oversample the dataoutput to determine if it's a '0' or '1.

    But I don't understand why you need this type of testing. I assume your final system will be packet based. What information will you get from the BER test you suggest compared to a PER test?
  • 0
    TI__Mastermind 36395 points

    Jay, 

    You cannot use SmartRF Studio for BER testing directly, I see two approaches.

    1. Use PER and estimate BER from it, this is something that is done often. If you choose a short packet of 4 byte preamble, 4 bytes sync word, 4 byte payload, and 2 byte CRC. All the bits in the sync word, payload and CRC must be correct for the packet to be good. Therefore the packet is 80 bits long and therefore to get 1% BER you will have failed packets in about 50% of the cases. So in many tests we estimate 1% BER by using a 3 byte payload and 50% packet error rate.
    2. Use transparent mode of the CC1xxx series devices and a R&S SMIQ signal generator or equivalent. The SMIQ can then take the output from the CC1x and compare it on the fly. This can be used to make BER testing directly.

    Regards,
    /TA

  • 0 in reply to TER
    Expert 1240 points
    TER,
    Not really sure that I do.
    Trying to validate/invalidate Carlson's Rule or TI's interpretation of it.
    As in getting back to RF engineering fundamentals versus implementation in Silicon.
    --
    Z
  • 0 in reply to TA12012
    Expert 1240 points
    TA12012,

    Why do your solutions alway cost more than my 1st house:-)

    --
    Jay Zebryk
  • 0 in reply to Zebryk
    TI__Mastermind 39860 points

    If you measure sensitivity for a given RX filter BW and a set of PHY parameters you are indirectly measuring Carson's rule (RX BW >= Signal BW). If you want to validate Carson's rule, which is simply an estimation of the signal BW, you should use a spectrum analyzer and put the radio into TX.

    To check if you are using the correct RX filter BW to account for PHY parameters and the crystal tolerance you simply measure PER (or BER) vs frequency offset. Assuming you use our development kit and SmartRF Studio you can put one radio in RX and the other in TX. The TX radio can be set to different frequencies to emulate the crystal frequency offset (initial tolerance, temperature drift, ageing). In order to increase the effective RX filter BW without increasing the noise bandwidth (the latter defines the sensitivity, = -174 dBm + 10*log(BW) + SNR + NF) you should refer to the Feedback to PLL Feature (FB2PLL) described in the user guide.