mfilt.firtdecim to dsp.FIRDecimator

Question

Alan Jayson 2019-9-11

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/479925-mfilt-firtdecim-to-dsp-firdecimator

回答： Alan Jayson 2019-9-20

fir_cic_compensation_32_bit_filter.m

I’ve used “mfilt.firtdecim” to generator a fir decimation by 2 cic compensation filter (At least 10 years) that import into Filter Designer app.

Want to set CoeffWordLength, InputWordLength, InputFracLength, OutputWordLength, OutputFracLength, AccumFracLength & OverflowMode.

I’m looking to dsp.FIRDecimator to do the same job but it not clear to me how to do that.

If anybody could point me in the right direction, I would appreciate it.

Alan

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Jyothis Gireesh 2019-9-18

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/479925-mfilt-firtdecim-to-dsp-firdecimator#answer_392434

在 MATLAB Online 中打开

Please make use of the following code snippet to set the above mentioned properties in dsp.FIRDecimator object.

Hd = dsp.FIRDecimator('Numerator',cfir,... 
                            'CustomCoefficientsDataType',numerictype([],32,24),'CoefficientsDataType','Custom',... 
                            'FullPrecisionOverride',false,... 
                            'CustomOutputDataType',numerictype([],32,0),'OutputDataType','Custom',... 
                            'CustomAccumulatorDataType',numerictype([],32,0),'AccumulatorDataType','Custom',... 
                            'OverflowAction','Saturate'); 

Since there is no name-value pair to set the customize the numerictype of the input you may have to typecast the input before passing it as input to the "Hd" object. You may make use of the following syntax to achieve the typecasting

inSignalConverted = fi(inSignal,numerictype(1,32,0));

Please go through the following documentation links on dsp.FIRDecimator and casting fixed point objects if you need any clarifications on the same

dsp.FIRDecimator - https://www.mathworks.com/help/dsp/ref/dsp.firdecimator-system-object.html

Cast fi objects - https://www.mathworks.com/help/fixedpoint/ug/casting-fi-objects.html

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

Answer 2

Alan Jayson 2019-9-18

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/479925-mfilt-firtdecim-to-dsp-firdecimator#answer_392439

dsp.FIRDecimator object (Hd) can't go into the Filter Designer application (There's an error), only DFILT/MFILT objects work like "mfilt.firtdecim" in the Filter Designer application (Import Filter object).

If MATHWORKS is going to kill "mfilt.firtdecim" I hope they make a converter.

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Jyothis Gireesh 2019-9-20

fir_cic_compensation_dspfir.m

Hi Alan,

I replaced the "mfilt.firtdecim" code with the corresponding code in dsp.FIRDecimator. But I was unable to replicate the error you mentioned above. Could you give further clarification regarding 'dsp.FIRDecimator object (Hd) can't go into the Filter Designer application'?

I am also attaching the edited code for your reference.

请先登录，再进行评论。

Answer 3

Alan Jayson 2019-9-20

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/479925-mfilt-firtdecim-to-dsp-firdecimator#answer_392768

filter designer error.jpg

Hi Jyothis,

Here's a picture of the screen with the error on the code that you attached.

After "filterDesigner(firdecim)" Filter Designer open you must import the object into Filter Designer (That's when you get the error, running the m code doesn't import the object).

Alan