CRC checker routine

[Grant]

Greetings all,

I am reading some RS-422 data that seems pretty spiky. Wild point
removal doesn't quite seem to fit the bill as there are often dropouts
that go to zero (or inside the range of what can be expected) instead
of +/- infinity. I was hoping that there was a crc routine that could
be accomplished in IADS as our hardware cannot supply that function
for us.

Best Regards,
Grant Lichtsinn

[Kathy Rodittis]

Hi Grant-

We don't have any CRC routines right now. But even if we did, without
the hardware's cooperation, I'm not sure it would do you any good
since you need something to "inject" the correct CRC within the data
stream before our software can use it to validate the incoming data.
Do you have a way to include a CRC in the original data source?

-Kathy

[Kathy Rodittis]

We have talked to Acra about the possibility of including this feature
in general for PCM data sources. If that happens then we'll definitely
be implementing the cooresponding CRC routines on our side as well.

Does checking the decom status help at all in determining when to
"ignore" the data?


-Kathy

> > Grant Lichtsinn- Hide quoted text -
>
> - Show quoted text -

[Pat]

HI Grant,

Take a look at our spike detection/removal technique that removes
spikes that occur within the exepcted range of data, here is a prior
Google post to get help you get going, also checkout the help system
for proper setup

http://groups.google.com/group/iads/browse_thread/thread/d55e339dd06fead5/bf37dcaba6f96c5c?lnk=gst&q=Spike#bf37dcaba6f96c5c

-Pat

> > - Show quoted text -- Hide quoted text -

[John Bretz]

If the "0.0" case is the only real problem and your data isn't about
0.0 normally, a derived equation might help...

John

[Grant]

Kathy, in fact the avionics people we are working with on this RS422
line had a good mind to put a 32-bit CRC at the end of their message.
I have yet to ask which CRC 'polynomial' their are using but I would
think that could be an input into the CRC checker feature. Because I
expect the problem to be noise in the 422 line that I am reading, I
wouldn't expect checking the decom status to be much help but I am not
a 'network' either. Since I am only listening to the RS422 line, I
would think a simple use last good value or user set value would
suffice for me much like you already have for wild point editing.

Pat, I will give the spike detection a shot in the mean time but would
prefer checking the CRC just in case spikes do occur in actual
recieved/verified data, or to clean up CAS messages where spike
detection wouldn't work.

Grant

On Dec 8, 6:21 am, Pat <p...@iads-soft.com> wrote:
> HI Grant,
>
> Take a look at our spike detection/removal technique that removes
> spikes that occur within the exepcted range of data, here is a prior
> Google post to get help you get going, also checkout the help system
> for proper setup
>

> http://groups.google.com/group/iads/browse_thread/thread/d55e339dd06f...

[ka...@iads-soft.com]

Ok Grant if you get a CRC in there let us know.



On Tue Dec 8 9:55 , Grant sent:

--

You received this message because you are subscribed to the Google Groups "IADS" group.
To post to this group, send email to ia...@googlegroups.com.
To unsubscribe from this group, send email to iads+unsub...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/iads\?hl=en.