Prev: Q1.15 calculation
Next: Demodulator - Slicing by float to fixed conversion
From: skyworld on 8 Jun 2010 09:28
Hi,
I'm new to matlab and I'm developing fft matlab code for hardware
prototype. I have passed floating point code and try to verify it with
fixed point algorithm. I met problem here.
1). I use fi(x) to define input and then use fft(x) to check matlab
original results. I got message: Function 'fft' is not defined for
values of class 'embedded.fi'. Why I can't use fft() here? Is there a
solution?
2) I use complex input X, use fi(X,1,16,15) to define its type and use
fi(wn,1,16,15) to define twiddle factor. after my matlab code I got
output Y. The problem is that I can't process Y. I tried to plot Y,
but with this warning: Warning: Imaginary parts of complex X and/or Y
arguments ignored. I then try to use abs() to get its module, i.e.
re^2+im^2, but failed with this error message:Function 'fft' is not
defined for values of class 'embedded.fi'. why this happens and how
can I solve it?
Can anybody give me some help? thanks
From: skyworld on 9 Jun 2010 01:13
On Jun 8, 9:28 pm, skyworld <chenyong20...(a)gmail.com> wrote:
> Hi,
> I'm new to matlab and I'm developing fft matlab code for hardware
> prototype. I have passed floating point code and try to verify it with
> fixed point algorithm. I met problem here.
> 1). I use fi(x) to define input and then use fft(x) to check matlab
> original results. I got message: Function 'fft' is not defined for
> values of class 'embedded.fi'. Why I can't use fft() here? Is there a
> solution?
> 2) I use complex input X, use fi(X,1,16,15) to define its type and use
> fi(wn,1,16,15) to define twiddle factor. after my matlab code I got
> output Y. The problem is that I can't process Y. I tried to plot Y,
> but with this warning: Warning: Imaginary parts of complex X and/or Y
> arguments ignored. I then try to use abs() to get its module, i.e.
> re^2+im^2, but failed with this error message:Function 'fft' is not
> defined for values of class 'embedded.fi'. why this happens and how
> can I solve it?
> Can anybody give me some help? thanks

Hi,

is there anybody could help me?
From: Jerry Avins on 9 Jun 2010 09:14
On 6/9/2010 1:13 AM, skyworld wrote:
> On Jun 8, 9:28 pm, skyworld<chenyong20...(a)gmail.com> wrote:
>> Hi,
>> I'm new to matlab and I'm developing fft matlab code for hardware
>> prototype. I have passed floating point code and try to verify it with
>> fixed point algorithm. I met problem here.
>> 1). I use fi(x) to define input and then use fft(x) to check matlab
>> original results. I got message: Function 'fft' is not defined for
>> values of class 'embedded.fi'. Why I can't use fft() here? Is there a
>> solution?
>> 2) I use complex input X, use fi(X,1,16,15) to define its type and use
>> fi(wn,1,16,15) to define twiddle factor. after my matlab code I got
>> output Y. The problem is that I can't process Y. I tried to plot Y,
>> but with this warning: Warning: Imaginary parts of complex X and/or Y
>> arguments ignored. I then try to use abs() to get its module, i.e.
>> re^2+im^2, but failed with this error message:Function 'fft' is not
>> defined for values of class 'embedded.fi'. why this happens and how
>> can I solve it?
>> Can anybody give me some help? thanks
>
> Hi,
>
> is there anybody could help me?

Many at comp.dsp use Matlab (I don't) but your questions would be better
asked in a Matlab forum. (There is one, but I don't know its name.)

Jerry
--
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������
From: dbd on 9 Jun 2010 12:41
On Jun 9, 6:14 am, Jerry Avins <j...(a)ieee.org> wrote:
> ...
> Many at comp.dsp use Matlab (I don't) but your questions would be better
> asked in a Matlab forum. (There is one, but I don't know its name.)
>
> Jerry

comp.soft-sys.matlab

Dale B. Dalrymple
From: Avier on 9 Jun 2010 13:26

the thing you can check out is that fi may be a defined name for some
function... use some other name ,,, i have used fft many a times and it
does not give problem ,,,also check if data type is right ,,,we have got
data type converter for that also in matlab ... the best way is to write
fft in help and check out the example for settings

there are complex plots also which you can use ,,, just go to
communication blockset >>>sinks (in simulink ) just check its name and
write it in help you will get the function name ,,,also yo can plot img and
real part seperately
for eg

i=img(Y);
r=real(Y)
plot(i):hold on
plot(r)

or use the abs function
ab=abs(Y)
plot(ab)

check the size of y,,,,(use size(y) command)
if it has more than one dimensions,, plot command cant work ,, start a
loop and store values in another variable (the required ones) and just do
the above mentioned commands


if it helps tell me ,,,,or if you get problems with loop !!!!!!
 |  Next  |  Last
Pages: 1 2
Prev: Q1.15 calculation
Next: Demodulator - Slicing by float to fixed conversion