Re: analysis software

So we'll probably use a tapped long-pattern pseudorandom shift

> register (64 or 96 bits maybe) and post-scramble to zap bit > crosscorrelations; XOR gates are free. That gives us random 16-bit > numbers with flat probability distribution. If we digitally lowpass > filter that, we'll get bandlimited Gaussian noise. If we clock that > whole system from the DDS synthesizer, we can set the noise bandwidth > 3db point anywhere we like it, mHz to MHz, without affecting RMS > amplitude, which is pretty cool.

You can't change a flat probability distribution into a Gaussian by low pass filtering.

Filtering has no effect on the probability distibution. Regards,

Mike Monett

Reply to
Mike Monett
Loading thread data ...

AoE, p 655, first line on the 2nd column. You might also google "central limit theorem."

Are you yet another nym for MassiveProng?

John

Reply to
John Larkin

Don't be stupid, John. The Central Limit Theorem doesn't talk about the frequency spectrum of noise, only about the probability distribution.

I don't have AoE, John. Please post the relevant section, or some other web url.

Regards,

Mike Monett

Reply to
Mike Monett

You don't need a book and you don't need a link; you only have to think for about 15 seconds.

John

Reply to
John Larkin

There is always a bit of degree snobbery around.

Thinking is pretty good and a bit of research. Dont take anything for granted.

Reply to
Marra

No, John. Mr. Monett easily predates MassivelyWrong. Please do not lump him in with the the auktards.

--
 JosephKK
 Gegen dummheit kampfen die Gotter Selbst, vergebens.  
 Click to see the full signature
Reply to
joseph2k

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^!!!!

Well, he did tell me that what we'd been working on all week was totally wrong, and that I should scan and send him a chapter from Win's book, and hunt down links for him. All without bothering to first check to see if he was right. Under those circumstances, I have a legal right to be a little crabby.

John

Reply to
John Larkin

OK. I am crabby all the time. I am trying to out crabby Lucy.

--
 JosephKK
 Gegen dummheit kampfen die Gotter Selbst, vergebens.  
 Click to see the full signature
Reply to
joseph2k

I did do a google search to check on your claim. There was utterly and absolutely no confirmation available. The best I could find was an old TEK AWG tutorial that showed averaging uniform probability distributions would tend to a normal Gaussian distribution, but very poor. To make a Gaussian, they used table lookup.

So if the trick were known, it should have been easy to find. But it's not available anywhere.

I finally got a copy of AoE, 2nd edition. Page 655 does indeed talk about low-pass filtering a PRBS pattern to generate pseudo-Gaussian noise. There is no derivation offered, no proof, and no links for reference. This is one of my main complaints about AoE. Lots of useful info, but no links to learn more. And the main reason I have eliminated all hard copy is - no text search.

The parts I do recall from AoE is the filtering works only up to about 0.1% of the main clock frequency. This means you have to generate a 1GHz PRBS pattern to get Gaussian noise up to 1MHz. That is a tremendous waste. There are much better ways of doing this.

For example, see the schematic on pp 662-663.

Any professional Gaussian noise source must pass rigorous proof that it is truly Gaussian. For example, look at the Box-Muller proof in Wikipedia:

formatting link

If you intend to sell high-value Gaussian noise generators to your aerospace customers, you need something a bit more substantial than a vague reference in AoE. There are too many good ones available to screw around with junk.

The real problem is you claim to have spent a week on this problem with your engineers, and you were put out because someone questioned your claims. This is extremely poor engineering. It only took me seconds to reject your approach once I found the information you were working with. It would not support your product claims.

What this really says is your engineers know that you hire at will, and they are terrified at loosing their jobs and homes. So they do anything you want and agree with any wild idea that passes through your head. We know about these - most of them end up in this newsgroup.

This leads to the spectacle of your canvassing the newsgroups and your friends for single-source parts from Maxim. Which everyone says to not use since they are an unreliable vendor. You knew this, but ignored all the available advise. You not only risked your customer's projects, you also risked your own company's survival on this single blunder. A competent engineer would have put a stop to that before it got to pcb. But they can't do that without losing their job.

Other enlightened errors you have shown us are forgetting to include parasitic suppression resistors on a simple pcb, which turned out to be too small to include them after the fact. If your engineers could do their job properly, that pcb should never have been made.

There are many other similar blunders you have documented here, and I don't have time to look them all up right now.

Your job is to ask why do these things happen, and to prevent them before they occur. You have not done this, and your customers can look forward to more amateur blunders that could jeopardize their programs. They don't need this.

Somewhere along the line, you have to give up the idea of hire and fire at will, and give your people long-term contracts that prevent you from firing them if they disagree with you. If they are any good, you owe them the confidence they deserve to do their work. And the freedom to tell you to go to hell when you get in their way.

Then, if they are any good, you will see the innovation and development you so desperately crave.

But it's not going to happen the way you are working now.

Regards,

Mike Monett

Reply to
Mike Monett

Untrue.

The Central Limit Theorem.

Not very difficult to derive it.

It depends on how well do you want to match the Gaussian distribution. If I am not mistaken in my coarse estimates, somewhat 1/30 of the initial bandwidth should give a reasonably accurate approximation (better then 1% RMS).

Truly Gaussian source must have infinite peak power.

The hypocrisy and bs will work as usual.

Vladimir Vassilevsky DSP and Mixed Signal Consultant

formatting link

Reply to
Vladimir Vassilevsky

The generator on page 662 of AoE uses a FIR filter with a cutoff of about 1/20 of the clock rate. But he's starting with a 1-bit stream, which has a dual-impulse PD, about as bad as it gets. My 16-bit random numbers start with a flat PD, which gives me a serious leg up. Summing even six samples will result in a pretty good Gaussian distribution.

The real problem isn't to get a decent Gaussian probability distribution; that's easy. The hard part is to get the spectrum right, and not fry the FPGA in the process.

If you sum a finite number of samples, the PD falls to zero on the tails, instead of trailing off to infinity. That allows you to get a reasonable amount of RMS signal through your amplifier stages. If you get crazy and try to generate a true normal distribution, the amplifier *will* clip, and the resulting distribution will be a lot worse.

I note here, for any individuals that didn't know it already, that lowpass filtering involves summing.

John

Reply to
John Larkin

AoE is a basic cookbook for amateurs and beginners. The recipes from there are fairly good, but may not be absolutely best for all occasions. And as every cookbook, AoE doesn't go into the depth.

That makes a lot of sense. For the Gaussian generators of this type, the rule of thumb is 12 samples.

It should not be a problem to generate a reasonably flat gaussian noise in a band by the use of FIR filter.

I note here, for those who don't know, that any filtering involves some sort of summing, and the lowpass filtering is not conceptually different from bandpass or highpass.

Vladimir Vassilevsky

DSP and Mixed Signal Design Consultant

formatting link

Reply to
Vladimir Vassilevsky

Mike M>I don't have AoE, John. Please post the relevant section,

formatting link

Reply to
JeffM

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.