FPGA Intellectual Property
PCI Express*, Networking and Connectivity, Memory Interfaces, DSP IP, and Video IP
Announcements
Intel Support hours are Monday-Fridays, 8am-5pm PST, except Holidays. Thanks to our community members who provide support during our down time or before we get to your questions. We appreciate you!

Need Forum Guidance? Click here
Search our FPGA Knowledge Articles here.
5917 Discussions

multichannel FIR output timing error

Altera_Forum
Honored Contributor II
826 Views

When I use multichannel fir filter, the timing of output channels is not correct. 

The output generated by 1 channel shifted. For example, First channel data filtered out to second channel, second channel data filerd out to third channel. 

Why is it incorect? Please help me.
0 Kudos
5 Replies
Altera_Forum
Honored Contributor II
89 Views

what version and what parameters? 

 

can you just stick an impulse into channel 0 and grab a SignalTap with the input and output?
Altera_Forum
Honored Contributor II
89 Views

Hi,pjw0112 

I have the same problem. Please give me some advice. Thanks! 

 

Guo
Altera_Forum
Honored Contributor II
89 Views

Did either of you find a solution to your problem? 

 

I have a two channel filter but have only been putting signals into channel 0. I have been debugging a DC block module. Eventually I decided there was a timing issue on the input (output of the FIR). With a bit of messing I found the signal appears on channel 1. 

 

Is this the same as what you are seeing? 

 

I have to make a few hardware modifications to my test boards before I can stuff a signal into channel 1. 

 

my dc block simulates ok. I use Altium Logic Analyser (which is the same as signal tap, i think). 

 

any advice? 

 

Quartus 9.2, 2 channel, decimating by 8, parrallel 

 

 

update- I have connected channel 0 and it seems that the output is swapped. That is, I is on channel 0 and Q is on channel 1, feeds the FIR. The output is Q channel 0 and I on channel 1. 

The actual filtering and data signals are fine. 

 

I have probably crossed my Is and Qs somewhere. I shall recheck the input to the FIR, but the output is definitly as above.
Altera_Forum
Honored Contributor II
89 Views

I have checked and checked and the inputs to outputs are reversed 

I with SOP & Valid, Q with EOP & Valid go in 

Q with SOP & Valid & Channel 0, I with EOP & Valid & Channel 1 comes out 

 

Anyone?
Altera_Forum
Honored Contributor II
89 Views

I think I've had a similar issue with a multichannel CIC filter. 8 inputs between SOP and EOP and these appear in a jumbled (but in a constant) position at the outputs relative to the output EOP/SOP. 

 

My problem was the thing generating the input data. 

There were some spurious DV's on the input, the first signal to enter the filter was an EOP rather than an SOP and the third packet had one DV too many between the SOP and EOP. 

 

Bad design on my part:oops:. I fixed these and problem dissappeared. I suspect it was the extra DV between SOP/EOP.  

 

Check your inputs carefully.
Reply