bessel filter

does any body know how to determine the analog prototype of a bessel filter???

thanks

bngguy wrote:

> does any body know how to determine the analog prototype of a bessel filter???
> 
> thanks

Eh? Do you mean, given a digital Bessel filter (there are few reasons to
want one), how can one determine the analog prototype from which it was
derived?

Jerry
-- 
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������

hi! Jerry,
          am trying to derive a digital filter from its analog prototype... 
let me know if u have any leads....

thanks


Jerry Avins <jya@ieee.org> wrote in message news:<4074c6e9$0$1658$61fed72c@news.rcn.com>...
> bngguy wrote:
> 
> > does any body know how to determine the analog prototype of a bessel filter???
> > 
> > thanks
> 
> Eh? Do you mean, given a digital Bessel filter (there are few reasons to
> want one), how can one determine the analog prototype from which it was
> derived?
> 
> Jerry

Try this:
http://www.crbond.com/papers/bsf2.pdf

(found by Googling for Bessel filter).

"bngguy" <loosepraje@yahoo.com> wrote in message
news:519f37cf.0404080750.2e62fa18@posting.google.com...
> hi! Jerry,
>           am trying to derive a digital filter from its analog prototype...
> let me know if u have any leads....
>
> thanks
>
>
> Jerry Avins <jya@ieee.org> wrote in message
news:<4074c6e9$0$1658$61fed72c@news.rcn.com>...
> > bngguy wrote:
> >
> > > does any body know how to determine the analog prototype of a bessel
filter???
> > >
> > > thanks
> >
> > Eh? Do you mean, given a digital Bessel filter (there are few reasons to
> > want one), how can one determine the analog prototype from which it was
> > derived?
> >
> > Jerry

bngguy wrote:

> hi! Jerry,
>           am trying to derive a digital filter from its analog prototype... 
> let me know if u have any leads....
> 
> thanks
> 
> 
> Jerry Avins <jya@ieee.org> wrote in message news:<4074c6e9$0$1658$61fed72c@news.rcn.com>...
> 
>>bngguy wrote:
>>
>>
>>>does any body know how to determine the analog prototype of a bessel filter???
>>>
>>>thanks
>>
>>Eh? Do you mean, given a digital Bessel filter (there are few reasons to
>>want one), how can one determine the analog prototype from which it was
>>derived?
>>
>>Jerry

That's the other way around from what I guessed. Why do you want to do 
that? It isn't usually worth it. We use analog Bessel Filters for their 
  nearly constant group delay in the pass band, sacrificing amplitude 
response to achieve it. Symmetric FIRs have perfectly constant group 
delay and are relatively simple to design, so why compromise? When an 
FIR filter is derived from an analog Bessel prototype, the performance 
is usually disapointing. The amplitude sacrifices still mar its 
performance and its phase response is poorer than the prototypes. Unless 
this is a homework assignment or an assignment from a boss you can't 
reason with, do something more sensible.

Jerry
-- 
Engineering is the art of making what you want from things you can get.
&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295;

"Jerry Avins" <jya@ieee.org> wrote in message
news:4075aa41$0$1650$61fed72c@news.rcn.com...
> bngguy wrote:
>
> > hi! Jerry,
> >           am trying to derive a digital filter from its analog prototype...
> > let me know if u have any leads....
> >
> > thanks
> >
> >
> > Jerry Avins <jya@ieee.org> wrote in message
news:<4074c6e9$0$1658$61fed72c@news.rcn.com>...
> >
> >>bngguy wrote:
> >>
> >>
> >>>does any body know how to determine the analog prototype of a bessel
filter???
> >>>
> >>>thanks
> >>
> >>Eh? Do you mean, given a digital Bessel filter (there are few reasons to
> >>want one), how can one determine the analog prototype from which it was
> >>derived?
> >>
> >>Jerry
>
> That's the other way around from what I guessed. Why do you want to do
> that? It isn't usually worth it. We use analog Bessel Filters for their
>   nearly constant group delay in the pass band, sacrificing amplitude
> response to achieve it. Symmetric FIRs have perfectly constant group
> delay and are relatively simple to design, so why compromise? When an
> FIR filter is derived from an analog Bessel prototype, the performance
  ^^^ (did you mean IIR?)
> is usually disapointing. The amplitude sacrifices still mar its
> performance and its phase response is poorer than the prototypes. Unless
> this is a homework assignment or an assignment from a boss you can't
> reason with, do something more sensible.

As noted above, usually, you only design IIR filters from analog prototypes, not
FIR.  If the OP is constrained to using an IIR (e.g. by available processing
power, need to adjust on the fly, etc.) but still wants "good" phase properties,
then an IIR Bessel filter designed from an analog prototype might be the answer.

I have also heard that the digital version loses some of the desirable
phase/group delay characteristics of the analog prototype, but I believe it's
still better than e.g. Butterworth in that respect.