Where to ask multi disciplinary questions?

Question

I want to find out about maximising the rate of entropy collection using an analogue to digital converter fed by a noise source. It involves Shannon entropy calculations, log normal distributions, ADCs and signal amplification /scaling. It's a bit of statistics, electronics and mathematics, perhaps even a derivative.

The problem is the keenness of some to immediately vote the question off topic because it crosses forum boundaries. With the ubiquity of micro controllers and system on a chip chips, this must be a common issue.

Where is the best place to ask this?

(I tried asking my actual question at stats.SE but no one could answer).

Answer 1

I find this interesting because I have had questions that involve more than one discipline and can't be pinned to any one forum. And then I read some of the advice that suggests you stick to one component of the problem and concentrate on that. That made me smile, because the end result of keeping it specific is going to be demands for more details. Ask about the A/D converter, and they demand to know what you're feeding into it. And then they want to know what you're doing with your sampled data. And so on.

Maybe the closest you can get it to ask each piece in each forum, and link the questions so that the curious can still get ahold of the other details. Otherwise, I'd say, just ask the questions that will get you the most information, and take the resulting abuse--it's okay as long as you get answers. You're never going to make everyone happy. Some will complain, and some will give answers. If you don't make a habit of it, it will just be another war story for your blog.

Answer 2

If you are asking about the whole thing, then you haven't done your homework and the question is too broad, probably on any site. Ask about the specific problems you are stuck on. These should be separate questions, each asked where they fit best.

Actually, if you have a lot of questions to ask, then again you haven't done your homework and are essentially asking us to design your project for you. It's your project, and you should be doing most of the design and figuring out the details. We here to help on this site with electrical engineering specifics you might get stuck on.

Answer 3

I think that was a perfectly legitimate question for EE.SE, since, for what I understand, involves the relationship between the (nominal/average?) voltage across the reference and the mathematical form of the lognormal distribution of samples from that reference (and then the math to obtain an entropy expression and maximizing that).

Disclaimer: although I got a faint glimpse of what the solution involves, this is not among my fields of expertise.

Sadly, I think most people answering or commenting probably didn't fully understand the problem and tried to convince you that you didn't understand your problem.

BTW, you didn't get downvotes, so it appears that most people deemed the question legitimate or even interesting. The lack of answer may be because it appears (and mathematically probably is) a difficult problem.

Maybe try to add some detail about what you already know, e.g.: some pre-digested math expressions adapted to your case; links to (freely available) references to the theory involved, additional info about the final goal of the project (if you can disclose them).

All this could (1) show that you really did all your homework (so people won't think you should know better) and (2) induce people to give a stab at the problem without the need to do all the preliminary research on the web.

Good Luck!

P.S.: if you ever find a solution by yourself, please come back and post it as an answer to your question. I'd really like to know!

Answer 4

I would ask on math.se, as it's an optimization problem. You should do more work on your own before asking, though, as it should be pretty vanilla. Estimate an equation for your distribution. Express it in such a way that it reads "entropy = ...", and take the derivative with respect to Vref. That's how optimization is done. If you can't do the "entropy =..." step, then your real question is "how do I do the entropy = step?".

Your limit of 1023 puts an unnecessary constraint on your problem, and makes the math harder. If your issue is your saturation-- roll up your sleeves, do the problem without the saturation, and ask how to extend your solution with the saturation in place.

Despite the fact that you've dismissed Olin's response as "what you expect of EE.SE", he has a valid point. The world doesn't owe you an answer, and a bit of grunt work on your part will make your question much better on whichever stack you choose to ask it on.

Answer 5

Seeing this topic is still getting attention after two weeks, I'll be more specific about what is wrong with the question you posted. Here is my thought process as I read your question:

I have a circuit that generates random noise which I've measured with a 10 bit ADC. The following is the sample distribution:

OK so far. The graph looks plausible for this kind of noise source.

You will notice that it's a log normal distribution

No, I wouldn't. Maybe some time in the dark past a particular probability distribution was given the name "log normal" in some class, but if so, that has been long forgotten. The OP has now alienated all those who haven't just taken the class where this was mentioned, are working in the theoretical parts of EE, or happen to have bumped into this recently enough to remember what the implications of "log normal" are. Surely some people will know, probably the more academic types, but that will be quite a small subset of users here.

Since he didn't define it, link to a reference (probably wouldn't have followed it anyway since it's his job to put the salient points here), or explained whatever part of it being log normal is relevant here, he just lost most of the audience.

The writing style also gives a hint of the "research paper" attitude. That's being technically right and rigorous, but difficult to follow unless one already knows what is being talked about. Mabye it isn't, but it often seems this is done deliberatly: "I'm smarter than you are because my math it more impenetrable than yours."

I'll just let this slide for now because it isn't clear how important this distinction is, and the relevant aspects might become clear from context later.

You can see that approximately 14,000 samples were at a value of 1023 or above

No, I can't. For reference, here is the diagram being referred to:

I can see that the frequency of occurance at 1000 is well below 5000, probably less than 1000 from eyeballing the graph. I do see a box with a top that could be at 14,000, but that box is not mentioned anywhere, and in any case is below 1023, not above it. What the ...?

Clearly I'm not getting something here. Probably best not to touch this question because I'll just look stupid. If this was written more accessibly and less arrogantly, maybe we could work it out, but not with this question.

they are effectively a source of random entropy

I can see the randomness, and the plot showing the relative probabilities of any particular value occurring, but what's the relevance of "entropy"? From context, this seems like some specifically defined way of measuring something about the randomness of the stream of samples, but like "log normal", whatever definition I might have understood from that has been lost in time.

This guy either has no clue about his audience, or is deliberately being academically arrogant to feel superior.

Shannon entropy rate per sample

It's definitely been too long since this was covered in a class (if it ever was) for me to remember what that means. Again, it seems like some kind of randomness measure.

The next paragraph explains that the point of the question is to maximize the randomness of the resulting samples by deciding what section of the overall probability distribution histogram to pick off. Why couldn't he just say that in the first place? He mentions scaling (which is really what changing the reference voltage does), but has ignored shifting the window around (adding a offset). That seems a obvious thing to do, but why isn't it considered? It is easy enough to do electrically. I would engage and ask about this if I didn't think I'd just get more Shannon entropy log blah blah babble in return.

Also, it's not really clear why not scaling and offsetting to pick off a narrow area around the hump of the curve isn't good enough. That will cause the A/D to clip to 0 and 1023 more often. These are really "no data" and therefore don't count as useful readings? What about the usual approach of using only the low bit of the A/D? If the window was chosen carefully so that the probability of clipping at 0 and at 1023 were equal, then the low bit should still be random and usable from the clipped samples too.

However, engaging with this guy feels like it's going to be painful, so I'll go find a bunch of simpler questions to answer in the time it would take going back and forth with this ivory tower weenie.

Also, if this guy has a equation for the probability distribution (I'm assuming that's what "log normal" refers to), and knows how to evaluate this Shannon entropy stuff he's on about, why doesn't he just write the equation of Shannon entropy as a function of the scaling factor and maximize it? How is that not obvious, especially considering the high-falutin theoretical stance he's taken? Something doesn't add up. It feels like penny-wise and pound-foolish. But, I don't feel like spending a lot of time and then just looking foolish myself as a result, so screw this.

Answer 6

You should pick a community which could potentially generate the most knowledge relevant to your question that you will be able to absorb and be willing to apply. For example, if you're an EE expert, chances are that you got the EE part of your problem just right, and people on EE.SE won't tell you much of that you don't know already. Conversely, if you're a complete EE dummy, it will be hard for you even to formulate the question correctly, and you may get answers which are either irrelevant or too generic. Or people will just tell you to go read a book. Or people tell you to use a particular schematic which you're unable to understand, so the answer is of no use.

In your case, if question is purely about a particular property (sample entropy) of a given distribution, you should really consider asking it on stats.SE. You don't even have to talk about the voltage there: just say that your distribution has a parameter you can control, and show how it factors in the distribution.

However, if you want to know what could be done electrically to improve your generator, you should ask here. I wrote an answer to your question and I hope it will be helpful to you and improves the generator beyond what would be possible with only optimizing Vref. But if you tell me that you're not going to change your circuit no matter what, I will be left wondering why did you pick this site to ask.

Answer 7

Stack Exchange operates on a simple rule. A question is welcomed at any page that the question is on-topic for, even if it may be more on-topic somewhere else. Migrations should only happen if the question is off-topic at the initial stack and on-topic at the other (and quality). Close votes should not be abused just because someone thinks its better off at another stack when its perfectly valid on the original one. IT'S OP'S CHOICE. Overlapping sites does not force migrations.