What is wrong in this IIR filter code?

Hi,

Given are following coefficients to a second order butterworth highpass 
filter with a 3 dB cutoff freq of 4.48 Hz:
double a0 = 1.0;
double a1 = -1.999722319818149;
double a2 = 0.999722358365939;
double b0 = 0.999861169546022;
double b1 = -1.999722339092044;
double b2 = 0.999861169546022;

that is used in the following direct form IIR filter implementation:

double Yn, Yn1, Yn2, Xn, Xn1, Xn2; // initialized to 0 before run

double Filter(double Xn)
{
Yn = (1 / a0) * (b0 * Xn + b1 * Xn1 + b2 * Xn2 - a1 * Yn1 - a2 * Yn2);

Xn2 = Xn1;
Xn1 = Xn;
Yn2 = Yn1;
Yn1 = Yn;

return Yn;
}

But the filter jumps to infinity very fast no matter what it is feed with.

Anyone who can see what I did wrong?????

Thanks
Ole 

Sorry - the coefficients are actually:
double a0 = 1.0;
double a1 = -1,990048023748777;
double a2 = 0,990097300073064;
double b0 = 0,995036330955460;
double b1 = -1,990072661910921;
double b2 = 0,995036330955460;

But the problem is still the same - hope anyone will be able to give me a 
hint.

thanks
Ole 

.... and the sample frequency is 4000 Hz. 

"Ole" <ole@blabla.com> writes:

> Hi,
>
> Given are following coefficients to a second order butterworth highpass 
> filter with a 3 dB cutoff freq of 4.48 Hz:
> double a0 = 1.0;
> double a1 = -1.999722319818149;
> double a2 = 0.999722358365939;
> double b0 = 0.999861169546022;
> double b1 = -1.999722339092044;
> double b2 = 0.999861169546022;
>
> that is used in the following direct form IIR filter implementation:
>
> double Yn, Yn1, Yn2, Xn, Xn1, Xn2; // initialized to 0 before run
>
> double Filter(double Xn)
> {
> Yn = (1 / a0) * (b0 * Xn + b1 * Xn1 + b2 * Xn2 - a1 * Yn1 - a2 * Yn2);
>
> Xn2 = Xn1;
> Xn1 = Xn;
> Yn2 = Yn1;
> Yn1 = Yn;
>
> return Yn;
> }
>
> But the filter jumps to infinity very fast no matter what it is feed with.
>
> Anyone who can see what I did wrong?????
>
> Thanks
> Ole 

It looks like you may have some numerical issues - your poles are
really, really close to the unit circle. Or you may have some bug in
another part of your program, e.g., something may be stomping on your
coefficients or data. I'd bet on the former, though.

Also, your cutoff appears to be much lower than 4.48 Hz.  Try moving
your cutoff frequency to something sane, like 10 Hz or so.
-- 
%  Randy Yates                  % "Bird, on the wing,
%% Fuquay-Varina, NC            %   goes floating by
%%% 919-577-9882                %   but there's a teardrop in his eye..."
%%%% <yates@ieee.org>           % 'One Summer Dream', *Face The Music*, ELO
http://home.earthlink.net/~yatescr

Ole wrote:

> Sorry - the coefficients are actually:
> double a0 = 1.0;
> double a1 = -1,990048023748777;
> double a2 = 0,990097300073064;
> double b0 = 0,995036330955460;
> double b1 = -1,990072661910921;
> double b2 = 0,995036330955460;

Are the commas in place of decimal points in your actual code? If so, 
the semantics of the comma operator will give you big-time nonsense 
there.


Martin

-- 
Quidquid latine scriptum sit, altum viditur.

Ole skrev:
> Sorry - the coefficients are actually:
> double a0 = 1.0;
> double a1 = -1,990048023748777;
> double a2 = 0,990097300073064;
> double b0 = 0,995036330955460;
> double b1 = -1,990072661910921;
> double b2 = 0,995036330955460;
>
> But the problem is still the same - hope anyone will be able to give me a
> hint.

Hi Ole.

Your name indicates you are posting from Scandinavia?
If so, be aware that comma is a local decimal separator
that the compiler might not recognize. Use dot instead. 

Rune

Rune Allnor wrote:
> Ole skrev:
>> Sorry - the coefficients are actually:
>> double a0 = 1.0;
>> double a1 = -1,990048023748777;
>> double a2 = 0,990097300073064;
>> double b0 = 0,995036330955460;
>> double b1 = -1,990072661910921;
>> double b2 = 0,995036330955460;
>>
>> But the problem is still the same - hope anyone will be able to give me a
>> hint.
> 
> Hi Ole.
> 
> Your name indicates you are posting from Scandinavia?
> If so, be aware that comma is a local decimal separator
> that the compiler might not recognize. Use dot instead. 
> 
> Rune
> 

One would think the compiler would complain.

John Sampson skrev:
> Rune Allnor wrote:
> > Ole skrev:
> >> Sorry - the coefficients are actually:
> >> double a0 = 1.0;
> >> double a1 = -1,990048023748777;
> >> double a2 = 0,990097300073064;
> >> double b0 = 0,995036330955460;
> >> double b1 = -1,990072661910921;
> >> double b2 = 0,995036330955460;
> >>
> >> But the problem is still the same - hope anyone will be able to give me a
> >> hint.
> >
> > Hi Ole.
> >
> > Your name indicates you are posting from Scandinavia?
> > If so, be aware that comma is a local decimal separator
> > that the compiler might not recognize. Use dot instead.
> >
> > Rune
> >
>
> One would think the compiler would complain.

Maybe, but I doubt gcc would.

Rune

"Rune Allnor" <allnor@tele.ntnu.no> wrote in message 
news:1165341730.208747.107760@n67g2000cwd.googlegroups.com...
>
> John Sampson skrev:
>> Rune Allnor wrote:
>> > Ole skrev:
>> >> Sorry - the coefficients are actually:
>> >> double a0 = 1.0;
>> >> double a1 = -1,990048023748777;
>> >> double a2 = 0,990097300073064;
>> >> double b0 = 0,995036330955460;
>> >> double b1 = -1,990072661910921;
>> >> double b2 = 0,995036330955460;
>> >>
>> >> But the problem is still the same - hope anyone will be able to give 
>> >> me a
>> >> hint.
>> >
>> > Hi Ole.
>> >
>> > Your name indicates you are posting from Scandinavia?
>> > If so, be aware that comma is a local decimal separator
>> > that the compiler might not recognize. Use dot instead.
>> >
>> > Rune
>> >
>>
>> One would think the compiler would complain.

Yep - the compiler did complain and I changed it, but the problem stilll 
existed. I have however located the error now - it was somewhere else in my 
code :-(

Hope taht anyone else will be able to use the example and sorry for wasting 
your time.

Thanks
Ole 

On Tue, 05 Dec 2006 13:34:01 GMT, Randy Yates <yates@ieee.org> wrote:

  (snipped)
>
>It looks like you may have some numerical issues - your poles are
>really, really close to the unit circle. Or you may have some bug in
>another part of your program, e.g., something may be stomping on your
>coefficients or data. I'd bet on the former, though.
>
>Also, your cutoff appears to be much lower than 4.48 Hz.  Try moving
>your cutoff frequency to something sane, like 10 Hz or so.

Hi Randy,
  Ole's tryin' to build a filter
whose cutoff freq is roughly Fs/1000 !!!

Isn't attempting to build such a "high-Q" filter 
using a simple 2nd-order filter just askin' for trouble?  

That filter doesn't seem to be realistic.
Maybe Ole should perform some lowpass filtering 
and decimation.

See Ya,
[-Rick-]