how do we end up with statements like 'millimeter waves can deliver typically between 10 and 50 GHz of bandwidth' getting published in ieee spectrum? are there no electrical engineers at the ieee anymore? apparently this article is by a 'tim hornyak' with a degree in journalism from carleton university in ottawa. what the fuck?
Is "between 10 and 50GHz" not a description of a range of widths of a band?
Are bands of that width not typically delivered with mm waves?
(There's more definitions of the word "bandwidth" than counting bits per second. It has, at least, uses in both RF and in data networking -- and the former use is predates the latter.)
10 to 50 gigahertz is a range of bandwidths, yes. that's not the problem
i don't know why you're bringing up bits per second. neither i nor the clueless loser who wrote the article were talking about bits per second
the quote doesn't say 'designs that deliver 10 to 50 gigahertz of bandwidth typically use millimeter waves'. it says the opposite: 'millimeter waves can deliver typically between 10 and 50 GHz of bandwidth'. that's clueless nonsense. it's not even wrong. bandwidth is delivered by a medium, not by a signal in it; the signal is closer to being what the bandwidth is delivered to. the signal occupies or consumes or has bandwidth, which is close to the opposite of delivering it. and a millimeter-wave signal can be of any bandwidth at all up to about 300 gigahertz, including—obviously, one would hope—bandwidths of well under a megahertz. (in theory you could transmit or detect a millihertz-bandwidth millimeter-wave signal, but that probably requires exotic instruments like an atomic clock.)
it's complete nonsense to try to describe the 'typical' bandwidth of a millimeter-wave signal. it's like talking about the typical mass of objects made out of atomic matter, or the typical distance traveled by photons. is jupiter or a tardigrade a more typical-sized atomic-matter object? it depends entirely on context. there is an upper limit (objects much larger than jupiter will stop being made of atomic matter pretty soon) and a lower limit (probably you need at least a dozen or so atoms of lithium before you have an 'object') but there are many orders of magnitude of slop within that limit
if we were talking about bits per second, it might make sense to talk about a signal delivering something, but even in that case the information capacity of the signal depends on many more considerations than just the wavelength. the bandwidth, for one, but also the relevant sources of noise, the transmit power, the path loss, and the efficiency of the coding scheme used. so you'd still have a lot of orders of magnitude of slop, and plenty of mmwave signals aren't even used for communication, so trying to characterize their shannon capacity is a somewhat questionable enterprise
It's not nonsense, though. It's just a generalization.
Here's another generalization: A car typically weighs between 1 and 3 tons, and typically travels on the highway between at speeds between 50 and 80MPH.
This generalization misses (many!) rather common outliers, and that's OK since it not meant to be particularly precise. That's the way of generalizations: They're generalized.
yes, i agree that it's reasonable to talk about the typical weight of cars. that is because they vary over many fewer orders of magnitude than the weights of objects made out of atoms or the bandwidths of submillimeter-wave signals, so it makes sense to talk about 'outliers' and edge cases such as golf carts and tanks. by contrast, there is no sense in which that is true of the bandwidths of submillimeter-wave signals, in part because they are mostly not of human origin, so your implied analogy is invalid
but, as i explained, that's not the only dimension along which the statement is nonsense; millimeter waves also don't deliver any hertz of bandwidth. it's a really multifaceted gem of cluelessness
the implication of the last line of your comment seems to be that you think i am not familiar with the site's guidelines, but it is unclear why. perhaps you think criticism of the ieee's editorial and human resources choices runs counter to the guidelines? perhaps you think there is nothing to be learned from my criticism? you'll have to be more specific if you want to have a substantive conversation
my criticism of the ieee is deserved, necessary, in the public interest, and, furthermore, substantial and informative. why does it bother you? is your sister editor of ieee spectrum or something? obviously you're not tim hornyak because if you'd graduated from journalism school you'd be in favor of robust, substantial criticisms of public institutions like the ieee when they're failing society like this, and you're not an electrical engineer either
