Light Speed Internet: Questions & Answers

[NetMage]

Hello, I know I'm just full of silly ideas, but afterall this is the physics forums, and I happen to have lots of questions :P. So, basically what I'm interested in, as the topic suggests, light speed internet. Now now, I know many of you would respond with, "its been done --fiber optics!". However, what about sending information on light waves rather than through fibers? What setbacks has there been? Does it all come down to when observed the wave collapses thus distorting the information?? Or have we simply not figured out how to modulize the information? I looked through here and a few other places before posting, but there were really no informative answers.

 

[DrChinese]

Hello, I know I'm just full of silly ideas, but afterall this is the physics forums, and I happen to have lots of questions :P. So, basically what I'm interested in, as the topic suggests, light speed internet. Now now, I know many of you would respond with, "its been done --fiber optics!". However, what about sending information on light waves rather than through fibers? What setbacks has there been? Does it all come down to when observed the wave collapses thus distorting the information?? Or have we simply not figured out how to modulize the information? I looked through here and a few other places before posting, but there were really no informative answers.

Microwave transmission towers - as well as satellites and fiber optics - are all examples of light (photons) used for signaling applications. It goes at c (actually a bit less).

 

[Dr Lots-o'watts]

...what about sending information on light waves rather than through fibers?...

It's the same thing. Fibers act as a guides ("roads") for modulated light waves.

 

[NetMage]

Hm, interesting. So would quantum computers be the answer to ridding lag or delays? It seems that satellite has about a 6 second delay due to it having to ping so many places, but I know for a fact that if it takes light from the sun ~8 min to reach the earth...that even if 'light speed waves' were traveling 6 different places, it should not take 6 seconds considering the distance being covered is far less than that of the sun. I can infer from this though that it must have something to do with the time it takes the computers to modulize and demod the information being sent. Would quantum computers have any impact on this problem?

 

[Faxmachinen]

Don't forget that ping measures the total time it takes both ways, and that the satellite isn't an ISP. Thus you get a predicted ideal ping equal to four times the distance to the satellite[1] divided by the speed of light. I calculate this to be approximately 479ms. And that is if you're directly beneath the satellite, and your ISP is next door.


[1] http://www.suite101.com/content/geosynchronous-satellite-orbits-a48472"

 

[timetravel_0]

NetMage, I think your missing the point of light(fiber optics) as a medium for data communications. Data is only transferred in small chunks called packets. From your PC a packets are sent to the modem which are then clustered into a larger packet, sent to a router on the ISPs network which is then clustered with other packets from other people on the network, which is then sent to another point clustered and so - eventually the packets are reduced back down once they reach the destination and then broken down by the modem of the server your trying to reach. These packets are sent at a specific rate. Higher rates need higher bandwidth.

What makes fiber optics a better medium is the way data is transmitted vs. in cable. Cable uses electricity where digital information is modulated into a wave and transmitted through the cable signal. The bandwidth for this technology is limited to the frequency of the modulated wave. Cable also has a limit on distance that data transmission is reliable, so this technology is limited. However, fiber uses light and it doesn't transmit data by modulated waves, but rather pulsating light. This means the only limit on bandwidth for a single fiber is the width of a photon - that would be very very very high speeds! However, computers can't even process or send data at that speed. Currently, fiber optics has yet to dominate over cable - in most situations fiber is converted down to cable, or phone line(DSL, and dial-up) and then is transmitted into your home - thus limiting the fiber technology even more. Currently fiber can transmit 320Gb/s(GigaBITS/Sec - that's 327680Mb/s!) which most homes getting cable only getting up to 12 - 25Mb/s and DSL is limited to about 1-12Mbs.

I'm sure quantum computers will revolutionize the internet once they are built and used worldwide. Just as Dial-up limited us to simple HTML web pages and corny MIDI music, Fiber brought about bandwidth intensive AJAX Technologies, and Video & Music Streaming. There are experiments where scientist have slowed light using Bose–Einstein condensates - which if used with current computer technology, we can transmit data in pulses down to the size of a photon and then send it out at the speed of light - allowing HUGE chunks of data Billions of Billions of Billions of Gigabytes - in an instant.(This could be an exaggeration... + or - some...haha) Technology then too would take a shift as we developed computers that can think like humans, that can make logical and emotional choices for us, what can think for us like a second mind. Searching the internet will be transformed from "Recipe for Bread" to "Will my aunt Susan like Wheat bread or White?" and the computers will be so smart they will know your aunt Susan better than you. Crazy, I know... but possible.

 

[Dr Lots-o'watts]

...So would quantum computers be the answer to ridding lag or delays?...Would quantum computers have any impact on this problem?

I'm not up-to-date on the possibilities of quantum computing, but you may want to read on optical computing. The concept is to replace electronic components of computers by optical equivalents. I personally had the opportunity to do research on all-optical switching, but to my knowledge, an all-optical computer has yet to be constructed.

Here's a starter: http://en.wikipedia.org/wiki/Photonic_computer

 

[NetMage]

Truley fascinating. So can any1 explain in detail what quantum computing really is? At the moment I have pretty limited knowledge on the subject as I have not taken the class yet =/ my physics is limited to calc-based Newtonian physics and what I've read in my free time.

 

[timetravel_0]

Quantum Computing will take advantage of quantum phenomena such as superposition and entanglement. The bit operation will be different that standard computing which is binary - on or off, 1s and 0s. Instead they will use the properties of the particle to describe data - giving them a wider range. Let's assume they can measure 8 different properties of a particle - that means they can move from binary to hex. In programming, each character is represented in a hex code. 00, 01, 02... A1, A2, A3... CD, CE, CF... and so on to FF. The programming language looks like:

x = $_POST["UserMessage"];
If(x = "Hello World") {
alert("I got your message, Hello to you!");
}

Which is then converted into Hex and looks something like:

AF FC 04 42 CD D8 78 DC 00 05 CC
CD 54 02 13 5D FC 04 42 54 02 55

The Hex Code in memory is represented not as Hex, but as Binary - so the code is then lengthened to:

0000 1111 1111 0011 0000 0010 0010 0001
0010 0001 0011 0100 0110 0101 0011 0000
0110 0010 0011 0000 0010 1010 0000 1111
0000 1111 1111 0011 0000 0010 0010 0001
0010 0001 0011 0100 0110 0101 0011 0000
0110 0010 0011 0000 0010 1010 0000 1111
0000 1111 1111 0011 0000 0010 0010 0001
0010 0001 0011 0100 0110 0101 0011 0000
0110 0010 0011 0000 0010 1010 0000 1111

and so on... Well with Quantum Computers - the code isn't transmitted in binary but rather hex(or something else.. don't know that much), making the bit much smaller - which is called a qubit(quantum bit). Smaller bits, more data within a given bandwidth. More Data means faster processing. Right now, data processing is limited to the size of an electron - with gaps in between, in the future with quantum computers - they are only limited to the number of combinations of properties for which a particle can be measured plus the side of the particle - which will have no gaps since it measures properties and not "heartbeats" pushing the electrons even closers... giving it a HUGE leap in processing power. The bit operation of quantum computer can be a bit more complex - so my example is only to understand it in a simple way.

Take the class, it will make more since of it.

 

[NetMage]

Yea I take QM I next semester, am very excited! As well as contemporary physics. Of your knowledge, do you know what universities or whom might be doing big research on quantum computers?

 

[Danger]

Microwave transmission towers - as well as satellites and fiber optics - are all examples of light (photons) used for signaling applications. It goes at c (actually a bit less).

And just where the hell have you been hiding, young man? I honest-to-**** thought that you'd bought the biscuit or at least retired. Nice to have you back.

 

[timetravel_0]

http://ctp.lns.mit.edu/research-quantum.html" to name a couple. I am sure most universities have quantum computing research centers since it is the up and coming technology. They are striving to perfect munipulation of the property of an electron without disrupting other electons - as well as methods for storing and transfering the data... It will be very complex...

If your interested in quantum mechanics I suggest you watch this video series on youtube called http://www.youtube.com/watch?v=bw5TE5o7JtE". It is a 15 video series and its quite interesting - its covers the basic principles of Particle Physics, Forces, The Standard Model, QM, and QED(Quantum Electro-Dynamics). Good Luck.

 

[NetMage]

Thank you, you've been quite helpful :) . And thanks again to the rest of you PF!

 

[Drakkith]

Interesting. I never realized how quantum computers worked. I've looked it up on wikipedia a few times, but trying to get all that into my head and understandable gives me a headache lol. So pretty much, the advantage of quantum computing is that instead of only 2 states to use, we can use more?

 

[russ_watters]

Hm, interesting. So would quantum computers be the answer to ridding lag or delays?

No, these are completely separate issues.

It seems that satellite has about a 6 second delay due to it having to ping so many places, but I know for a fact that if it takes light from the sun ~8 min to reach the earth...that even if 'light speed waves' were traveling 6 different places, it should not take 6 seconds considering the distance being covered is far less than that of the sun.

6 seconds would be an awful lot. Geostationary satellites are at something like 22,000 miles, which means they add about half a second to a ping (with 4 trips, as pointed out above).

I can infer from this though that it must have something to do with the time it takes the computers to modulize and demod the information being sent. Would quantum computers have any impact on this problem?

No, that is done extremely quickly - miliseconds - it is not an issue at all.

 

[Borek]

However, what about sending information on light waves rather than through fibers? What setbacks has there been?

If you take "light" literally, you don't have a reliable transmission. Rain, birds, fog - whatever you put between transmitter and receiver, it can get blocked any time.

But it has been done on many occasions, a friend of mine designed a "light phone" back in seventies, IIRC he planned to use a LED and a telescope for transmission. These were pre-digital times, and it was not easy to get a phone line in Poland. Not sure if he finally built a prototype or not.

 

[timetravel_0]

Interesting. I never realized how quantum computers worked. I've looked it up on wikipedia a few times, but trying to get all that into my head and understandable gives me a headache lol. So pretty much, the advantage of quantum computing is that instead of only 2 states to use, we can use more?

Well I had an entire post typed up to explain this and I pressed the backspace key when the text box was not in focus - and the browser backed me up a page... and i lost everything i typed... stupid IE...

Anyway. Yes, and no. A normal computer doesn't measure states - it reads pulses. Pulses of electrons(thousands or even millions of them) to represent a 1(on) or a 0(off). A quantum computer will measure the probability of the state of a single electron. I've done a bit more research on this and I THINK I get what its doing. An 8 qubit system - represented by 2³ will use 3 binary digits represented by a(000), b(001), c(010), d(011), e(100), f(101), g(110), and h(111). Each binary digit represents the probability that the electron is pointing up(1), or the probability that it is pointing down(0). I THINK this is what it is saying... I could be wrong.

Basically it is measuring different states of an electron... 8 different states for a 2³ qubit system... 16 for a 2⁴. Take for example the way computers process characters. A character is represented by a number - for instance Z is = 90. How do I know this... well I'm an ex programmer... but you can hold down the Alt key and type 90 on the number pad and get the letter Z when typing in a textbox or notepad. Whole numbers are not processed - but rather HEX (0,1,2,3,4,5,6,7,8,9,A,B,C,D,E,F). 90 in HEX is 5A - why? Because if you were to count in sequence to 5A you would have counted 90 different numbers. Hex gives the ability to store up to 256 different numbers into two digits. Hex is then processed by the computer in Binary, 1s and 0s. 0000 = 0, 0001 = 1, 0010 = 2, 0011 = 3... so on. So. 5A is represented in two bytes(2 sets of 4 bits): 0101 1010 = 5A = 90 = Z. The machine has to pulsate out 0 1 0 1 1 0 1 0 - which it does quite extremely fast...(It does BILLIONS of these a second!) but you have to understand that the path that the character Z takes to get from your keyboard, to the motherboard, to the processor, to memory, to your video card, then finally to your display - well is quite a task - but it does it instantly almost. But quantum scientist are hoping to do this more effectively(silly huh?) - instead of using thousands(or millions) of electrons in pulses to represent a single digit - they are hoping to use 1 electron to represent thousands(or millions) of digits!

I hope this answered your question...

 

[DaveC426913]

Well I had an entire post typed up to explain this and I pressed the backspace key when the text box was not in focus - and the browser backed me up a page... and i lost everything i typed... stupid IE...

Argh! I hate that!

(But it's not endemic to IE.)

 

[timetravel_0]

I can infer from this though that it must have something to do with the time it takes the computers to modulize and demod the information being sent. Would quantum computers have any impact on this problem?

No, that is done extremely quickly - miliseconds - it is not an issue at all.

I disagree, Quantum computers will make all the differents. Fiber has the abbility to TRANSFER at 320Gbits/s with current techonology. With Quantum Computers - or even Quantum Optics(Quantum Interent!) things will be INSAINLY Fast! Think in http://en.wikipedia.org/wiki/Yottabyte" /s!

 

[russ_watters]

You're confusing processing and data transmission. They are not the same thing.

 

[timetravel_0]

You're confusing processing and data transmission. They are not the same thing.

Na, I know I saw that typo awhile ago. I ment to use the word "transmit" and used "process" instead - was too lazy to change it :)

P.S. - The word "Yottabyte" didn't make you giggle?? Hmmm, Intellegent and Quick Master Yottabyte indeed is He!

 

[DaveC426913]

Na, I know I saw that typo awhile ago. I ment to use the word "transmit" and used "process" instead - was too lazy to change it :)

No, Russ is still correct. Quantum computers are about processing power, they are not about transmission speed.

A quantum computer could do more with what you give it, but it would still transfer data at the same rate as any other computer.

 

[timetravel_0]

No, Russ is still correct. Quantum computers are about processing power, they are not about transmission speed.

A quantum computer could do more with what you give it, but it would still transfer data at the same rate as any other computer.

I'm quite sure your misunderstanding my point - sorry. Currently, fiber optics though fast, is limited by the equipment used to transfer the data. I could be wrong, but I'm sure fiber has the potential bandwidth in the Terabits/s range. But current routers, switches, and servers are in the Gigabit/s range. Thus causing a bottleneck effect on the internet. Its amazing to think of having a Gigabit connection - however, most webservers won't transfer at that rate so you'll never realize the full potential. Quantum Computers will most likely first be used by large corperations. Companies like NetFlix will be able to provide all its users with ultra-high definition movies that download litterally instantly.(pending the user has a connection bandwidth that can handle those speeds). Can you now see my point?

Also, Quantum Internet will change everything - using entanglement to litterally teleport data across the world instantly. I can imagin the day when instead of shipping merchandise we will just download it and have a machine that fabricates it like a printer prints pages... I'm sure this would be impractical - but cool nonetheless. I can see that technology being useful in space mining and exploration. Imagin sending data of a schematic instantly to a mining station on an astroid in space where a fabrication unit can take that schematic and build the prototype on the spot using the materials mined - all without the need of local human interaction. I have some crazy ideas up in this noggin... *knock knock* :)

 

[russ_watters]

Also, Quantum Internet will change everything - using entanglement to litterally teleport data across the world instantly.

Heh, I was thinking that might be where you were going. No - again - quantum computing is about processing, not about data transfer. You cannot teleport data across the world instantly: that would be a violation of Special Relativity. The fastest you can get data from one place to another is the speed of light. And again, ping and throughput are entirely different things. You can have fast ping and slow throughput or vice versa depending on a bunch of different factors.

Also, you're out of date with the capacity of fiber:

Nippon Telegraph and Telephone Corporation have also managed 69.1 Tbit/s over a single 240 km fibre (multiplexing 432 channels, equating to 171 Gbit/s per channel).[20] Bell Labs also broke a 100 Petabit per second kilometer barrier (15.5 Tbit/s over a single 7000 km fibre).[21]

http://en.wikipedia.org/wiki/Optical_fiber#Optical_fiber_communication

My home internet speed is limited to 5megabits by my ISP, but I'm not sure there iis any compelling need for additional speed right now - and what am I transferring that is that big anyway? If it takes me two hours to watch a movie, why do I need to download it instantly? And if it is 20 gigs of blu-ray and my hard drive writes at 50 MB/sec, the fastest I can usefully download it is 400 sec or 400 megabits.

Certainly there are infrastructure needs - upgrading the world to those kinds of capacities - but currently available technology offers a pretty wide-open data path.

 

[timetravel_0]

Heh, I was thinking that might be where you were going. No - again - quantum computing is about processing, not about data transfer. You cannot teleport data across the world instantly: that would be a violation of Special Relativity. The fastest you can get data from one place to another is the speed of light...

I think I've had that point pushed into my brain several times now within this topic... I understand that quantum computing is about processing and not about data transfer. My point is in relation to the server user to the end user, not just the end user. What I'm trying to point out is the relationship between processing power and data transfer effectiveness. Example:

I have a server - the specs - A quadcore processer - each core running at 3.0GHz in 64bit. That means it can send a 64bit thread of data 3221225472 times a second through each core. Thats only 768Gb/s - Let's say I had a 1 Terabit Connection - Hmm... can't seem to use it all... Let's say I host a large 50GB file that people want to download - and I have a million users trying to download the file at once. I can only allocate about 0.78Mb/s to each user - that means the users will only be downloading at 200Kilobytes/s - however it would me much less because of the processing power needed to establish and maintain each connection, the data process of the hard drive and the memory along with all the services running on my server. Realistically it would be in 100Kilobytes/s or less. 50GBs would take 150+ hours to download. This would occur even if the user themselves had a Terabit connection. In order to handle those users I need more servers.

However, with a quantum server that can process data in the Petabits/s, or even Exabits/s - as you can see from my example above - the one single quantum server can now handle those 1 million users, enabling the download to transfer instantly(probably only a couple of seconds or less...) for the users with a terabit connection.

Now as far as the Quantum Entanglement thing and the Internet - well I just briefly read something on it - you should google it, I could be wrong.

...My home internet speed is limited to 5megabits by my ISP, but I'm not sure there iis any compelling need for additional speed right now - and what am I transferring that is that big anyway? If it takes me two hours to watch a movie, why do I need to download it instantly? And if it is 20 gigs of blu-ray and my hard drive writes at 50 MB/sec, the fastest I can usefully download it is 400 sec or 400 megabits.

I've been running on a 24Mb connection for some time and if downloads are not in the 2 - 4MB/s range then I grow impatient. 5Mb/s wouldn't reach even 1MB/s downloads. You say you don't see a compelling need for additional speed - well. Let's go back 10 years ago. I had a 166MHz Processor with about 16MB of memory(maybe less...) I had the great old dial-up which got 1 to 2Killobytes/s on a good day. I'm sure many others were in the same position with the same equipment and speed. I remember getting into music - at the time music was in WAV format - which if you wanted good quality music, your file size would be 50MBs or more - my hard drive was only 80MBs(maybe less, can't remember). About this time, MP3s were the latest thing - I was a pioneer in pirated music for sure. It allowed audio files to have good quality at small file sizes, 1 - 2MBs. Even on dialup this took hours to download one song - even then the quality was still not CD quality. Forget video - videos were in poor quality, only 10 seconds long and had a resolution of about 128 x 128... In the last 10 years we went from that to the ability to download full quality music albums in seconds and watch high quality 1080p feature length videos instantly on the web.

Maybe we won't need faster internet - but if things follow this trend we will. And to make my final point - If you used a single fiber optic strand and used a simple laser to transfer data - the MAX bandwidth would be 3 Petabits/s.(Based off a 1mW Laser with a 620nm opening = 3 x 10¹⁵ photons/sec) I'm sure they use more strands and lasers that emits more electrons and I'm sure the calculation is a bit more complex... My point is, the bandwidth capabilities of fiber is, well... limitless. However - computer networking systems and their processing power do in fact control the speed of which the data can be transferred across that fiber... Am I making sense now??

 

[f95toli]

However, with a quantum server that can process data in the Petabits/s, or even Exabits/s - as you can see from my example above - the one single quantum server can now handle those 1 million users, enabling the download to transfer instantly(probably only a couple of seconds or less...) for the users with a terabit connection.

It doesn't work that way. Quantum computers are very good at certain problems, but no better (probably worse) than classical computers for others. No one has -as far as I know- been able to come up with a way to speed up communication links using a quantum computer (and not for lack of trying). I doubt there would be much point in trying to build e.g. a router using a quantum computer.
Quantum computers and classical computers use very different types of algorithms and are good a different things. The point is that quantum computers are NOT just "fast" computers; they are very different.