Strictly speaking, that's not true either. It is certainly possible to conceive of a purely analog system in which the transmission involves only a small number of discrete possible values. An analog video interface in which the video source can only do, say, eight levels of gray is one such example. In short, while most analog systems ARE capable of continuous variation between values, there's no real requirement that this must be so. It's hard to imagine a real need that would drive such a "quantized analog" scheme, but it surely could be implemented.
well
Actually, the heart of the definition of "digital" lies just where you might expect to find in - in just what that root word "digits" means. A "digital" system is one in which information is encoded as what (for lack of a better term) might be viewed as "numeric" values Ii.e., "digits") directly, as opposed to being represented in "analogous" form by some other quantity (e.g., the "voltage for sound pressure" example I gave earlier).
My favorite example, which I've often used in classes on this topic, is as follows:
Suppose you're supposed to record the temperature at a certain location at various times during the day. You might do this in either of two ways. First, you could get yourself a piece of graph paper, and every time you take a temperature reading, make a point on the graph and draw in the next line segment to connect it to all the previous points. Or you might simply write down your readings in the form of a table:
The graph paper is an "analog representation" in this model - the line varies just like the temperature did, to within the limits of how often and how accurately you took your readings (and could draw the line!). The table is a "digital" representation - information stored as symbols that represent numbers directly, but which have no real quality that varies in a manner akin to the original parameter.
-------------------- Not for me. Everything down to this quoted info was excellent. But I view the signal as digital for the following.......
Well sent CW is a conbination of dots, dashes, and a couple types of spaces. The dot period is the standard from which the dashes, and the couple types of spaces are constructed. I don't recall off-hand the numbers, but this is close enough for discussion:
dot = 1 unit interval (1 bit) dash = carrier on for 3 dot intervals. space between letters = carrier off for 4 dot unit intervals space between words = carrier off for 7 dot unit intervals
nothing is sent that is not a construct of THE dot unit interval, aka one bit.
The code is sent using a non-return-to-zero (NRZ) code.
Clock is recovered by the mind by simply hearing the signal and syncing up with the overall rhythm
"Bob Myers" wrote in message news:3f284e4d$ snipped-for-privacy@usenet01.boi.hp.com...
I agree with this also. I have seen systems that did use analog channels to carry specific signal levels.
Yes, and digits are exact and discrete representations. And if I make any electronic system, I can show that it is all analog, but by convention (and clever design), we can label a small number of well defined states, and make sure the hardware cannot (under normal circumstances) produce anything other than those states. We agree on the convention that (for example) in TTL logic, anything below
0.6 volts is a zero and anything above 2.4 volts is a one. If we measure the actual voltages present, we are likely to see just about anything from 2.4 to 5 volts present on an output when a logical one is produced. Depends on the exact logic family and the loading, but we agree on those boundaries. Likewise, we also see anything from 0.6 volts down to 0 volts as a logical zero. We have assigned the digits ourselves, and we accept that the hardware is digital because it is designed not to dally about at some halfway state. Of course, we have also built specific "guard bands" into the levels to be certain that no such signal is made. A 1.5 volt level is likely to be interpreted as anything if applied as an input, but it will not be produced by the system unless something is broken. Now, if we built logic that uses a four valued system, we would set four well defined intervals, along with "guard bands" to isolate them and be certain that they will not be confused with each other. Three little voltage windows will be "no man's land" here, and our four logic states would likely be 0, 1/3 of supply, 2/3 of supply, and the supply. Making it monotonic like this helps us to decide what an errant signal should be interpreted as- and the key is interpretation. If I were presented with a signal that showed only 5 discrete signal levels or values throughout, I would have no problem in accepting that as being a digital representation. If I can reproduce it with a small handful of states, and each is clearly unique, then it exhibits "digital-ness". If, however, it shows stretches of continuous value, then it is an analog signal that happens to have stretches of digital type data. We know that television is analog, but nothing prevents us from using that timing and framing to encode and store digital data. In fact, it has been done. Somebody made data backup recorders from VCRs a few years back. You could store the data as video frames. Here they were clearly storing digital data in an analog medium. But so does any of the old audio frequency modems.
Once again, I agree with you. Now, at the point where you decide what the closest integer value to your reading is, you have quantized your data. Yes, the graph paper is an analog representation, and yes, the table is digital. Each has a completely different "flavor" to it. But nothing (other than the problem with noise creeping into the system) prevents us from using a 100 valued system. And there are those ISD audio storage and platback chips which actually break the audio into one of 256 discrete levels and save it in a flash chip of sorts. These pretty much straddle the border between analog and digital because they have enough resolution that they are right at the noise limit for what you can tell apart in terms of "byte" value for each sample. They are designed to be digital in some respects, but clever work has made them store roughly sampled audio. I suppose the best way to put this whole thing is that we label those systems by convention, and we use the criterion of "ability to represent the contents in a small number of discrete states". As an aside, I find no problem with requiring Morse to be known to get a license. We can't reasonably claim it is too difficult- look at all the programmers who know most of the ASCII code table. And my kids can each remember about a hundred Pokemon cards with no problem- the difficulty argument is a hollow one.
But I never claimed otherwise- I clearly said it was a digital system. Just keep in mind that digital is not the exclusive property of "binary". ANY system that defines a set of discrete levels and disallows all else is digital, because, by definition, it can be represented using nothing more than digits. You can have any number of logic levels in a digital system- 2, 3, 5, 9... as long as the individual states cannot be confused with each other, your system is still useful. We limited ourselves to binary just because the logic was easy to make. It has often been suggested that trinary would be somewhat more friendly because the three states can carry more information in a given representation. But practical design and use of trinary computers is still a problem, and conventions rule.
----------- Erroneous. In that case, anyone near it wouldn't either, so there wouldn't be enough qualified personnel to operate it either, so it evens out. The EMP mythology is a lie which is *FAR* too widely propagated. Anywhere EMP would render equipment unusable, it would also kill the personnel and destroy everything else close at hand as well by blast. Those very levels of energy density go together intimately! No such thing occurs as seen in cheap sci-fi movies where magically everything solid state becomes unusable.
-Steve
--
-Steve Walz rstevew@armory.com ftp://ftp.armory.com/pub/user/rstevew
Electronics Site!! 1000's of Files and Dirs!! With Schematics Galore!!
http://www.armory.com/~rstevew or http://www.armory.com/~rstevew/Public
That bit seems to have largely been forgotten by today's "black box" operators, whose interest seems solely to lie in operating equipment that they have never even taken the covers off.
Radio is a branch of electrical engineering, which is a branch of applied physics.
--
Then there's duct tape ...
(Garrison Keillor)
nofr@sbhevre.pbzchyvax.pb.hx
Things changed so you could buy increasingly fancy equipment, which in turn made it easier for people not interested in technical matters to get into the hobby, which gave enough people to want simpler entry requirements, which reinforced the number of non-technical people in the hobby.
Let's make the test simpler --> let's have licenses that don't require a code test --> let's drop the code test completely.
And after a certain point, the next "logical" step after this is a desire to drop the technical test completely.
On one hand, people are lamenting that amateur radio (and other technical hobbies) can't compete in a world of cellphones, internet, and video games. Yet, too many think the solution is to try to compete, ie promote amateur radio as a communcations utility, when in reality it isn't nearly as reliable for that as any of the common devices available.
Promote it as in the old days, and maybe you won't get so many newcomers, but you will get a better quality of newcomer.
Mind you, I don't think amateur radio does a good job of promoting it. Too much effort has gone into reducing the entry requirements, and not enough of being visible so the kid who is interested in technical things will find it.
I'm not sure if it's perception or not, but it seems like the local astronomy clubs here are making an effort to promote their events beyond those already in the know. Public telescope parties were an annual thing when I was a kid, but up until, there was a long period when I was never seeing anything about this.
Fred posted: and technical investigations carried out by amateurs That bit seems to have largely been forgotten by today's "black box" operators, whose interest seems solely to lie in operating equipment that they have never even taken the covers off. >>
----------- But let's not let the FCC and ARRL know that. I believe that they believe that as long as they have Advanced, Amateur Extra, and other "grades" of license beyond General, all is well with the world.
More importantly, you need to know what to do about it if you _haven't_ got a clean, legal signal. To most, these days, I suspect, that means a trip to the shop.
--
Then there's duct tape ...
(Garrison Keillor)
nofr@sbhevre.pbzchyvax.pb.hx
By necessity, in many cases. The "clean, legal signal" example may not be the best, as final amplifiers and such may still be generally serviceable by the knowledgeable amateur. But if pretty much anything else in the transceiver goes bad, what are you going to do? It's not like the local Radio Shack is going to have the custom DSP chip or whatever you need in stock, even assuming that you had the equipment and skill needed to get fine-pitch surface-mount parts in and out of the thing...
Yes, the "boxes" are probably designed by a team of engineers where one guy designs the front end, another the output end, and some other guy does the digital stuff, and someone else the power supply, etc. There's probably one guy in charge who knows how everything fits together, but not too many details of individual sections. Makes it sort of hard for any one person to fully understand the magic box.
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.