best method for implementing an SRC

Hi,

  i would like to know which is the best method for implementing a integer
sample rate converter (decimation /interpolation) in terms of quality.
the conversion rates required are (2,4,8,16) both decimation and
interpolation.i had tried with polyphase implementation and lagrange
interpolation.lagrange interpolation is too spiky and polyphase resulted
in flat signal without images. But i would like to know what is the method
that is generally followed in implementing an SRC and the 
advantages/disadvantages of different methods if you guys have some
idea..
thanx in advance...

regards,
srikanth konejti

On Feb 1, 8:00 am, "srikk" <srikanth_...@rediffmail.com> wrote:
> Hi,
>
>   i would like to know which is the best method for implementing a integer
> sample rate converter (decimation /interpolation) in terms of quality.
> the conversion rates required are (2,4,8,16) both decimation and
> interpolation.i had tried with polyphase implementation and lagrange
> interpolation.lagrange interpolation is too spiky and polyphase resulted
> in flat signal without images. But i would like to know what is the method
> that is generally followed in implementing an SRC and the
> advantages/disadvantages of different methods if you guys have some
> idea..
> thanx in advance...
>
> regards,
> srikanth konejti

I think this is commonly done with FIR filters in the time domain. It
can also be done in the frequency domain with an OLA or OLS structure.

John