Is attenuation the right word to describe the loss of data transfer speed over long distances?

sort by: active | newest | oldest
Claude Shannon, in the 1940s described the rules that limit the dataRATE in a channel,against signal loss and noise. It is complex stuff to understand, but Shannon differentiates between DATA and INFORMATION - a subtle distinction which is worth finding more about.

If you think of a standard text file, and think how much it can be compressed without loss of words, you may begin to see the distinction: What you have BEFORE compression is data, what you have LEFT is pure information. Perhaps you can also see that corruption of the compressed file destroys information much more than corruption in the "originak fila becus u ca stll unnnerstan i"

Shannon was one of the 20th Century's giants of computing theory - up there with Turing, Mauchley and Williams.

Great stuff
Yes it is. :)
thegeeke5 years ago
Attenuation is when you have less signal at the destination than you do at the origination point. Although in a Ethernet environment that normally means the result is slower speeds and or loss of data, in other environments (such as analogue wireless audio), it would mean you are loosing information (yes, information, not really data...), but the speed is still the same.

So... basically... yes.
frollard5 years ago
distance doesn't affect speed directly per se...

You have signal, noise, and attenuation from start to finish.

Signal is how much power you put in one end
Noise is anything that shows up at the other end you didn't send.

The signal to noise ratio is very important for most physical protocols because over a certain threshold the receiver won't know what is being sent any more.

Many other minor things like delay (time between being sent and being received), crosstalk (hearing other things talking, like noise but looks like useful data), signal bounce (echo) cause other problems...

Oversimplified example:

Attenuation is the DROP in signal strength from start to finish. If I talk at 50dB 1 meter from you, you receive at near 50dB. If you are 100m away, that 50dB looks a lot more like 10dB when it gets there (arbitrary numbers). That 10dB is also mixed in with 20dB of say...wind and traffic noise...at which point, you can no longer distinguish what I'm saying.
In order to get my message to you, I need to
a) talk louder, this degrades the signal from the getgo generally because its harder to shout accurately, but it makes it go farther. "I AM OKAY!" shows up as "i an okae"
b) talk slower, more carefully, perhaps repeating myself to give redundant check-data. "I am good, okay is what I am, confirming, I am ok". If you listen carefully, you get part of the message and you know the entire meaning.
c) use better conductors to get the message across; talk over a string/cup phone, or use a real telephone. This can introduce delay and echo, crosstalk etc.

Different methods of data transmissions suffer in varying degrees to these problems. Careful design limits the effects of transmission loss but it happens.

...which is a long-winded "yes".
lol, exactly.
seandogue5 years ago
It *can mean that under certain circumstances.

Generally though, attenuation is used to describe the characteristic droop in amplitude due to dispersion or absorption to the space (and objects) around the source and inbetween the source and receiver, respectively.

"speed" would come into the equation only if the expected pattern was recognized by multiple valid "hits" in a binning system (for instance, spectral data is often statistically correlated over many non-simultaneous samples, which leads to an obvious issue of time), so "speed" could be measured by the threshold of N ~identical data points over time T for a valid communication. Otherwise, speed and attenuation don't really correlate.

Attenuation usually refers to a loss of signal to noise ratio.

Distance ~generally affects speed only by virtue of the frequencies at which the communication is sent and the intervening distance between those points.

I'll leave the rest up to the philosphers
I thought of another way to voice a part of my meaning

you can have a source located in the same room with a receiver and have attenuation that leads to a loss of data. And you can use a laser to losslessly send information to the moon only limited by its speed and that of the distance between in the amount of time it takes to get there.