I understand that there are 1’s and 0’s (on/off) in binary code to transmit data, but does the light inside the fiber optic flash billions of times per second to simulate the 1’s and 0’s? Im also curious if copper ethernet cables work in a similar way to transmit data.
In: Technology
The light flashes (or, rather, is sent in pulses). The light is converted into electricity on the receiving end, with the intensity of the light resulting in higher voltages. From the perspective of the receiving computer, nothing really changes. In the end it gets high and low voltages as it would with a copper wire.
The fiber to your house is just intensity modulated, on=1 off=0. More valuable fibers are [wavelength division modulated](https://en.wikipedia.org/wiki/Wavelength-division_multiplexing ), essentially different “colors” of light blinking different bits of the message.
Basically yes, the light flashes on an off billions of times per second. The actual data is encoded in a very sophisticated manner to reduce the actual flash rate but in the most fundamental sense, it’s still ‘Morse code.’
Ditto on the copper…
Basically, yes. You flash the light, and that’s information.
What’s important about fiber optics is that they can have a lot of signals going through them at the same time, without mucking each other up. The first way is to put in multiple colors of light, with a splitter for each color at the end. More advanced than that, you can control which direction the light is waving, and get a bunch more signals *for each color.*
So while each individual signal is relatively easy to generate, the combination of all of them is a LOT of data. A new record was set earlier this year of a **peta**bit per second, through one piece of glass.
>does the light inside the fiber optic flash billions of times per second to simulate the 1’s and 0’s?
Kind of
Full blown flashing is inefficient, it only lets you send 1 bit per change so sending information at 1 Gigabit per second requires adjusting the state of the light every nanosecond. The flashing is very resistant to noise which is good for longer distance, but is an inefficient use of a cable.
Copper ethernet cables basically do this but rely on multiple wires to give you better bandwidth. They have a pair of wires where one is high and the other low for a 0 and low/high for a 1. Similar to above this isn’t very efficient which is why gigabit ethernet has to use all 4 pairs of wires in the cable, so each one only has to flip every 2 nanoseconds at most.
The better way to ram data through your wire is with a form of modulation. What if instead of having your little colored laser alternate between full on and full off, it went between off, 1/3 on, 2/3 on, full on. Now you can slip 2 bits into each state change! This is Amplitude Modulation.
Now take a red laser and a blue laser, they can both travel in the same channel without interfering, and amplitude module both of them. Now each state change you can have 1 of 4 red levels and 1 of 4 blue levels for 16 options (4 bits), this is QAM-16(Quadrature amplitude modulation – aka two signals being amplitude modulated together). Still using the same cable, but now we can get 16x as much through it.
Boost your transmitter and receiver precision so they can differentiate 16 different levels per color, now each transition carries 1 of 256 different options so you can get 8x as much data down the line as you could just having one laser turning on/off. This is handy because it can let you greatly upgrade the bandwidth of an existing fiber optic cable just by adjusting the equipment on the end and doesn’t require laying potentially hundreds of miles of new cable just for more bandwidth.
That is exactly right for most optical fiber data transmission.
A modern 10 Gbps fiber network connection works by flashing a laser on-and-off 10 billion times per second. On means a binary 1, and off means a binary 0.
Typically a fiber network connection is 2 fibers. Computer A connects it’s laser to fiber 1, which then runs to computer B’s light detector. Computer B connects it’s laser to fiber 2, which then runs to computer B’s light detector.
It is possible to combine the upstream and downstream connections onto a single fiber, if computer A and computer B use different color lasers (e.g. computer A has a red laser and green detector, and computer B has a green laser and red detector). This is often how “fiber to the home” works – the laser/detector combos are more expnsive due to more complicated lenses/mirrors needed to connect a detector and a laser onto a single fiber, but it saves money on the fiber.
For even faster network connections, like 400 Gbps fiber connections, simply flashing the laser on and off isn’t good enough. A 400 Gbps connection will typically use “pulse amplitude modulation” and multiple lasers. Pulse amplitide modulation means that the laser isn’t just turned on or off, but it’s brightness is adjusted.
At 400 Gbps, PAM4 is used, which has 4 brightness levels – Off to transmit 00, 33% to transmit 01, 66% to transmit 10, and 100% to transmit 11. The laser is pulsed 50 billion times per second, allowing 100 gigabits per second to be transmitted with one laser. 4 separate lasers are then used to transmit the data in parallel. These can either be 4 standard color lasers, each connected to a separate fiber, running to a receiver with 4 separate detectors (This means a 400 Gbps fiber connection is often 8 fibers, 4 each direction). Alternatively, for long distance “wavelength division multiplexing” can be used – where there are 4 lasers of slightly different colors, connected to the same fiber, and 4 detectors sensitive to each laser color at the other end – this allows 400 Gbps over a standard 2 fiber connection.
Latest Answers