Sign in
Search imagedsp
Subscribe to imagedsp
imagedsp by Keywords
Ads
Discussion Groups
Discussion Groups | Image Signal Processing | Polyphase filtering for video scaling
Technical Discussions related to Image Processing (image coding, compression, digital effects, mpeg, etc)
Polyphase filtering for video scaling - peace_for_thriving - Dec 5 14:36:00 2005
Hi!
I am designing a 'P' Phase 'T' tap polyphase filter to implement video
scaling.
The way I am doing this is by designing a P*T tap filter in MATLAB and
then splitting it into 'P' pahses 'T' taps each.
However the contraint on this design is that each phase should have
unity gain at DC so each phase should sum up to unity.
However when I try to do this by normalizing each phase individually,
i screw up the over all frequency response of the polyphase filter.
Is there a better way to do this?
The goal is to design a polyphase filter meeting certain frequency
specs and have each phase summing upto unity.
Many thanks,
Peace Thriver
(You need to be a member of imagedsp -- send a blank email to imagedsp-subscribe@yahoogroups.com )
Re: Polyphase filtering for video scaling - Ranjith Parakkal - Dec 6 4:37:00 2005
Peace,
Are u sure the phases dont add up to the same value
automatically. If the filter design is ok, I think
this should happen by itself.
You have to normalize all the phases by the same value
though.
Regards
Ranjith
--- peace_for_thriving <peace_for_thriving@peac...>
wrote:
> Hi!
> I am designing a 'P' Phase 'T' tap polyphase filter
> to implement video
> scaling.
> The way I am doing this is by designing a P*T tap
> filter in MATLAB and
> then splitting it into 'P' pahses 'T' taps each.
> However the contraint on this design is that each
> phase should have
> unity gain at DC so each phase should sum up to
> unity.
> However when I try to do this by normalizing each
> phase individually,
> i screw up the over all frequency response of the
> polyphase filter.
> Is there a better way to do this?
> The goal is to design a polyphase filter meeting
> certain frequency
> specs and have each phase summing upto unity.
>
> Many thanks,
> Peace Thriver
(You need to be a member of imagedsp -- send a blank email to imagedsp-subscribe@yahoogroups.com )
Are u sure the phases dont add up to the same value
automatically. If the filter design is ok, I think
this should happen by itself.
You have to normalize all the phases by the same value
though.
Regards
Ranjith
--- peace_for_thriving <peace_for_thriving@peac...>
wrote:
> Hi!
> I am designing a 'P' Phase 'T' tap polyphase filter
> to implement video
> scaling.
> The way I am doing this is by designing a P*T tap
> filter in MATLAB and
> then splitting it into 'P' pahses 'T' taps each.
> However the contraint on this design is that each
> phase should have
> unity gain at DC so each phase should sum up to
> unity.
> However when I try to do this by normalizing each
> phase individually,
> i screw up the over all frequency response of the
> polyphase filter.
> Is there a better way to do this?
> The goal is to design a polyphase filter meeting
> certain frequency
> specs and have each phase summing upto unity.
>
> Many thanks,
> Peace Thriver
(You need to be a member of imagedsp -- send a blank email to imagedsp-subscribe@yahoogroups.com )
Re: Polyphase filtering for video scaling - R S - Dec 6 18:42:00 2005
Hi Ranjith,
I dont see a reason to assume that this would happen automatically and I dont observe this behavior either.
Can you elaborate on what you mean by the filter design being ok?
Many Thanks,
PT
Ranjith Parakkal <ranjithmails@ranj...> wrote:
Peace,
Are u sure the phases dont add up to the same value
automatically. If the filter design is ok, I think
this should happen by itself.
You have to normalize all the phases by the same value
though.
Regards
Ranjith
--- peace_for_thriving <peace_for_thriving@peac...>
wrote:
> Hi!
> I am designing a 'P' Phase 'T' tap polyphase filter
> to implement video
> scaling.
> The way I am doing this is by designing a P*T tap
> filter in MATLAB and
> then splitting it into 'P' pahses 'T' taps each.
> However the contraint on this design is that each
> phase should have
> unity gain at DC so each phase should sum up to
> unity.
> However when I try to do this by normalizing each
> phase individually,
> i screw up the over all frequency response of the
> polyphase filter.
> Is there a better way to do this?
> The goal is to design a polyphase filter meeting
> certain frequency
> specs and have each phase summing upto unity.
>
> Many thanks,
> Peace Thriver
(You need to be a member of imagedsp -- send a blank email to imagedsp-subscribe@yahoogroups.com )
I dont see a reason to assume that this would happen automatically and I dont observe this behavior either.
Can you elaborate on what you mean by the filter design being ok?
Many Thanks,
PT
Ranjith Parakkal <ranjithmails@ranj...> wrote:
Peace,
Are u sure the phases dont add up to the same value
automatically. If the filter design is ok, I think
this should happen by itself.
You have to normalize all the phases by the same value
though.
Regards
Ranjith
--- peace_for_thriving <peace_for_thriving@peac...>
wrote:
> Hi!
> I am designing a 'P' Phase 'T' tap polyphase filter
> to implement video
> scaling.
> The way I am doing this is by designing a P*T tap
> filter in MATLAB and
> then splitting it into 'P' pahses 'T' taps each.
> However the contraint on this design is that each
> phase should have
> unity gain at DC so each phase should sum up to
> unity.
> However when I try to do this by normalizing each
> phase individually,
> i screw up the over all frequency response of the
> polyphase filter.
> Is there a better way to do this?
> The goal is to design a polyphase filter meeting
> certain frequency
> specs and have each phase summing upto unity.
>
> Many thanks,
> Peace Thriver
(You need to be a member of imagedsp -- send a blank email to imagedsp-subscribe@yahoogroups.com )