Re: LF: Smart noise cancelling?!?

[Andy Talbot <andy.g4jnt@gmail.com>]

Not sure what you're saying, Alan

 By "narrow band filtering"   I actually DO mean the digital filtering applied in the FFT process.  It was the ability to go down to 30 milli-Hz bandwidth and lower  that showed Peter the bursty impulses etc were being filtered to something that passed the tests for being Gaussian

Back then the digitisation and first stages of decimation had to be done on external Motorola DSP cards, communicating with a PC for display and further slower decimation and filtering.

Programming  DSP chips in assembler was great fun then - far more interesting than doing it in high level code for soundcards

I still have three Motorola 56002EVM modules - doubt they're worth anything now ;-)    Blowing the 27128 EEPROMS they use to store the code is not that easy to do now - wonder if anyone kept their old EEPROM programmers?

Andy

On 3 November 2016 at 20:47, Alan Melia <alan.melia@btinternet.com> wrote:

Hi Andy things have changed since then. Real time FFTs were just beginning to be possible but not many had much experience at that time. I remember the cleverest stuff I saw was from a graduate student. It closed down some more traditional work, like an adaptive modem in weeks. Guessing from the comments on the latest modes now I think there seems to be little left to achieve in efficiency.

 

Most of us were still thinking in analogue ways then :-))  "Just 'cos it digital doesn't mean its better" I heard a number of times....we were to be converted !

 

Of course a well respected physicist opined in 1900 that there was little new left to discover in the field. Fortunately for me and the rest of the profession, he was slightly wrong :-)) 

 

Alan

G3NYK

 

 

 

 

----- Original Message -----

From: Andy Talbot

To: rsgb_lf_group@blacksheep.org

Sent: Thursday, November 03, 2016 7:45 PM

Subject: Re: LF: Smart noise cancelling?!?

But surely, when narrowband filtering is in place - as any narrow band mode will of necessity be doing internally - any wideband non Gaussian or bursty noise when applied to this narrow filter will eventually become Gaussian IN THAT BANDWIDTH

We first ealised this in teh original QRSS tests with G3PLX back in the 1990's.   73kHz was full of spikes and 'crud' from teh then existing Decca signals and other stuff.   But when Peter examined the output from the narrow filters, (the FFT bins)  it lookdd like, and appeared to show itself to be Gaussian.   He said it passed the tests for Gaussian noise

A mathematician could probably prove that any random non Gaussian signal if filtered sufficiently narrow in comparison to itsnature, would end up Gaussian in the filtered bandwidth. 

In fact, to end up non-Gaussian, it would have to have definite components repeating at rates very close to the reciprocal of the bandwidth of the filter.

Andy  G4JNT