• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 142 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.

ADC12J1600: FIR filter details

Oddbjørn Norstrand
Prodigy 130 points
Part Number: ADC12J1600

Hi

In the data sheet 7.3.6.3 it is stated that the output of the FIR filters are 15-bit complex. Is that the output of the final FIR filter or is that the output of each individual filter? How many bits are used between the filters and how are bits removed, dropped, rounded, truncated or saturated?

  • 0
    TI__Guru* 91947 points

    Oddbjorn,

    The 15-bit complex data is the final output of the final FIR filter. I am looking into your other questions.

    Regards,

    Jim

      

  • 0 in reply to jim s
    TI__Guru* 91947 points

    Oddbjorn,

    The resolution varies depending on the filter and decimation mode. Each decimation mode uses a cascade of 2 or more decimation filters.  The table below shows the resolution of each filter in the chain, and its decimation (M).

     

    Total Decimation

    1st filter

    2nd Filter

    3rd Filter

    4th Filter

    5th Filter

    4

    M=2, 16-bit output

    M=2, 15-bit output

    n/a

    n/a

    n/a

    8

    M=2, 16-bit output

    M=2, 16-bit output

    M=2, 15-bit output

    n/a

    n/a

    10

    M=2, 16-bit output

    M=5, 15-bit output

    n/a

    n/a

    n/a

    16

    M=2, 16-bit output

    M=2, 16-bit output

    M=2, 18-bit output

    M=2, 15-bit output

    n/a

    20

    M=2, 16-bit output

    M=2, 16-bit output

    M=5, 15-bit output

    n/a

    n/a

    32

    M=2, 16-bit output

    M=2, 16-bit output

    M=2, 17-bit output

    M=2, 15-bit output

    n/a

     

    At the output of each filter, saturation is performed.  Rounding also occurs, and the rounding is convergent rounding with ties rounded toward the nearest even output value (to avoid introducing a DC bias). 

    Regards,

    Jim

     

  • +1 in reply to jim s
    TI__Guru* 91947 points

    Please use this table instead. The other had an error.

     

    Total Decimation

    1st filter

    2nd Filter

    3rd Filter

    4th Filter

    5th Filter

    4

    M=2, 16-bit output

    M=2, 15-bit output

    n/a

    n/a

    n/a

    8

    M=2, 16-bit output

    M=2, 16-bit output

    M=2, 15-bit output

    n/a

    n/a

    10

    M=2, 16-bit output

    M=5, 15-bit output

    n/a

    n/a

    n/a

    16

    M=2, 16-bit output

    M=2, 16-bit output

    M=2, 18-bit output

    M=2, 15-bit output

    n/a

    20

    M=2, 16-bit output

    M=2, 16-bit output

    M=5, 15-bit output

    n/a

    n/a

    32

    M=2, 16-bit output

    M=2, 16-bit output

    M=2, 17-bit output

    M=2, 18-bit output

    M=2, 15-bit output