Phil wrote:
> Hi,
>
> I'm hoping that someone out there may be able to help me solve this
> problem.
>
> I'm trying to create a decimate by 2 filter that will have 4 parallel
> inputs and 2 outputs.  The reason for doing this is that I am
> receiving data into my FPGA at 4 times the clock FPGAs clock rate.
> This is being done by bringing in 4 samples per clock.
>
> I was thinking of using a polyphase filter, but this does not seem
> feasible for my 4-in 2-out scenario.  I think this has become a
> parallelism problem.  I'm trying to figure out an implementation that
> works, but if anyone has some advice for my, I would really appreciate
> it.
>
> Thanks in advance for any help.
>
> Phil

It turns out that I was being dumb when looking at this
implementation.

Instead of the line-up that I mentioned below, I will be using a
parallel NCO for downconverting the signal and then some half band
filters for bandlimiting and downsampling.  This is much less
probelmatic and I think it reduces the complexity of my design.

Phil

Hi,

I'm hoping that someone out there may be able to help me solve this
problem.

I'm trying to create a decimate by 2 filter that will have 4 parallel
inputs and 2 outputs.  The reason for doing this is that I am
receiving data into my FPGA at 4 times the clock FPGAs clock rate.
This is being done by bringing in 4 samples per clock.

I was thinking of using a polyphase filter, but this does not seem
feasible for my 4-in 2-out scenario.  I think this has become a
parallelism problem.  I'm trying to figure out an implementation that
works, but if anyone has some advice for my, I would really appreciate
it.

Thanks in advance for any help.

Phil