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.

SNR input parameter format in BCP for PUSCH

Hi,

I have question related to BCP input in TI 6670 Evm for Uplink LTE. My question is :

In the BCP example code for lte uplink there is an input (file) for SNR even and SNR odd.I assume this is SNR value for even and odd sub-slot of a sub-frame.I want to know what is the unit of this SNR even and odd parameter. Is it only the ratio of signal and noise or is it in decibel (dB) ?

Thanks and Regards

Piyush

  • Hi Piyush,

    Welcome to the TI E2E forum. I hope you will find many good answers here and in the TI.com documents and in the TI Wiki Pages. Be sure to search those for helpful information and to browse for the questions others may have asked on similar topics.

    Please check whether the document given in the below thread helps.
    e2e.ti.com/.../1493480

    Regards,
    Shankari
  • Hi Shankari,

    Thanks for the quick response.

    I went through the thread which you mentioned above. There is document for PUCCH_Receiver in which I was not able to find any information about SNR inputs. Then there is link for the latest MCSDK for Keystone I devices(C6657/70/78) user giude. There also I was not able to find any information about this SNR input format. And this thread which you shared has been raised by my collegue only. I had already discussed with him about this topic, but he was also not sure about the input format.
    Actually this SNR input is for PUSCH channel data part.

    Regards
    Piyush

  • Hi Piyush,

    OK, thanks for going through the docs and replying to us.

    Your post has been taken to an expert's notice. You may hopefully get reply from him.

    Thanks for your patience until then.

    Regards,
    Shankari
  • BCP_NoiseScalingFactors.pdfHi Piyush,

    The SNR parameters for SSL are computed usingf the one-dimensional noise variance which is what is read from a text file in the example. Please look at the attached document which explains the BCP noise scaling factors and the corresponding calculation given in the BCP PDK example. Hope this helps.

    Regards

    -Nitin

  • Hi Nitin,

    Thanks for the sharing the pdf on noise scaling factors. We are looking into it. Will come back to you if we have some doubt.

    Regards
    Piyush
  • Hi Nitin,

    After going through the "BCP_NoiseScalingFactors.pdf" document which you shared, a doubt has cropped in my mind.
    As per my understanding of the document the variables snrEven and snrOdd are equal to (1/noiseVariance) of even and odd slots of a sub- frame (LTE FDD) respectively. So let us suppose that a signal at Rx of eNodeB that is devoid of noise component. Then in this case the value of noise Variance should be 0. So if its value becomes 0 then values of snrEven and snrOdd becomes infinity(1/0 form). So if that is the case then what value will be fed to BCP in case of 0 value of noise Variance because infinite value cannot be practically fed.

    Regards
    Piyush

  • Hi Nitin,

    Can you please help us  with the following query :

    Piyush Kumar said:

    Hi Nitin,

    After going through the "BCP_NoiseScalingFactors.pdf" document which you shared, a doubt has cropped in my mind.
    As per my understanding of the document the variables snrEven and snrOdd are equal to (1/noiseVariance) of even and odd slots of a sub- frame (LTE FDD) respectively. So let us suppose that a signal at Rx of eNodeB that is devoid of noise component. Then in this case the value of noise Variance should be 0. So if its value becomes 0 then values of snrEven and snrOdd becomes infinity(1/0 form). So if that is the case then what value will be fed to BCP in case of 0 value of noise Variance because infinite value cannot be practically fed.

    Regards
    Piyush

    Also one more query regarding the polling and interrupt mode. In the Uplink file of example project for LTE polling mode has been used. For the Downlink file of the example project for LTE interrupt mode has been used. But when I tried to use a common interrupt mode configuration for both Uplink and Downlink I was receiving 4294967295 packets from BCP though I should have received only 2 packets.So my question is can the Uplink processing of BCP not work with interrupt mode configuration ?

    Regards

    Piyush

  • Hi Piyush,

    If the signal is coming from a real RF, it will always have some noise and therefore the noise variance cannot be zero. Anyway, you cannot supply an infinite value in the configuration registers so you can saturate the Noise scaling parameters to the lowest level according to the calculated noise variance.

    For the polling vs interrupt, there is no limitation on the BCP's part regarding different packet types with interrupts. The goal of the examples is to show both methods of processing packets i.e. using polling and interrupts.

    A simple approach to process multiple packet types (LTE-PDSCH, LTE-PUSCH, etc) is to mark the different packet types using the PS_INFO (protocol specific info) field in the BCP TM local header. The PS_INFO field is pass through and is recovered in the Navigator descriptor. In the BCP output processing function (called by a semaphore posted from the accumulator ISR), you can separate out the output packet types on PS_INFO and call the appropriate application level function to perform further processing on the particular packet. 

    I would request you to create a new thread for the polling vs. interrupt question or any other different questions you may have. Thanks.

    -Nitin