Kitz Forum

Internet => Interesting Websites => Topic started by: renluop on September 24, 2014, 01:30:39 PM

Title: Found item on noise data
Post by: renluop on September 24, 2014, 01:30:39 PM

Just poodling I found an item on noise data averaging.  Here (http://www.cirrusresearch.co.uk/blog/2013/01/noise-data-averaging-how-do-i-average-noise-measurements/) is a link just in case it might be of interest to anyone.

Title: Re: Found item on noise data
Post by: JGO on September 24, 2014, 05:20:25 PM

I'm wondering what is the point of the average described,  in relation to xDSL   ? 
Most instrumentation will measure the RMS of Gaussian noise accurately, and any other waveform will be in error. If this waveform error is constant you can compare results at different times, or average them OK, BUT averaging line noise and a switching PSU's noise say, is rather a waste of time.
Or am I missing something ?
 

Title: Re: Found item on noise data
Post by: NewtronStar on October 11, 2014, 10:11:27 PM

One of stats called FEC's from my observation's is a great way into seeing physical noise on the line if your interleaved, more noise higher FEC's less noise lower FEC's.

You will find most people with medium FEC errors are on the longer line whilst others with very high FEC errors the noise is being introduced onto the line by internal or rare exterior devices.

So yes you could use the stats FEC count as the noise data averaging.