On 17/11/22 14:22, Jerry Friedman wrote:
> On Wednesday, November 16, 2022 at 7:22:46 PM UTC-7, Peter Moylan
> wrote:
>> On 17/11/22 07:23, Jerry Friedman wrote:
>>
>>> Where I just noticed another Americanism, this one a symptom of
>>> the English preposition disappearance disease:
>>>
>>> "Graduated Munich University of Applied Sciences where he
>>> studied Communications Design."
>>>
>>> (But I think an American would still say "Communication
>>> Design".)
>
>> In the context of a university degree, the word Communication(s)"
>> is highly ambiguous. In an electrical engineering degree, a first
>> communications subject would cover topics like amplitude
>> modulation, frequency modulation, single sideband, and so on. An EE
>> majoring in Communications would follow this up with a subject
>> dealing with information theory, coding and decoding, and topics
>> like that. Somewhere in the stream you might also find the design
>> of transmitters and receivers, antenna arrays, near-field solution
>> of Maxwell's equations, noise analysis, and so on.
>
> Fiber optics? Fractal antennas? Whatever the hell cell networks do
> to give your call its own frequency? Cryptography? (I assume
> that's different from your "coding and decoding".)
I've been out of the discipline too long to know what is taught now.
Fibre optics certainly, but I'm not sure about the rest. Cryptography
is, I imagine, considered to be part of computer science rather than
part of EE, even though I'm very much involved with it at present.
Coding and decoding is basically about error-correcting and
error-detecting codes. (At the most elementary level, a single parity
bit is an error-detecting code, but of course we can do better. For an
example of something more complicated, look at the error-correcting code
used on CDs.) There's an overlap in concept with cryptography, in that
there are both block and stream coding methods. The difference is that
we're not trying to hide the information.
I was never comfortable with optical fibres when I was teaching that
stuff. With metal waveguides you can solve for the fields by direct
solution of Maxwell's equations, and from there you can deduce the
existence of various propagation modes, calculate their velocities, and
so on. In a fibre you have a cylindrical rather than rectangular
geometry, and the icing on the cake is variable refractive index. That
puts the mathematical analysis into the "too hard" basket, and you're
reduced to hand-waving explanations of what is going on.