Hi,
I am currently studying computer networks and I have been reading about how data is transmitted via copper wires and fiber optics.
I understand the very basics of computers, data, logic gates and binary code.
However, currently I am struggling to wrap my head around how data can be transmitted by copper/light? It just does not make sense to me!
Can someone shed light on this?
In: Technology
Remember that every single object in the universe has electrical properties. Some things have greater values for them, some of them smaller, to the point that we don’t notice them.
* inductance
* capacitance
* voltage
* resistance
We pretty much characterize everything based on those electrical properties. So, cabling systems and hardware are designed in such as way as to send a signal. The systems are designed to work within specific constraints of those properties. Examples of this that are easy to see:
* Limits on cable distance
* Twisted pairs in wiring
* Separation space for communications wiring
* voltage levels of the actual cables
The characteristics of those physical properties I mentioned above are the reasoning for this. Now, lets talk in general about signals. This is generally applicable to digital copper transmission systems (coax networks like cable use frequency modulation, which is different).
The network hardware will sit at some voltage. It varies based on the equipment, but it can often be at 0V. A cable connects two pieces of network hardware electrically. Signals are sent by applying a voltage to that cable relative to the 0V reference. That is just a fancy way of saying they will apply -5V, or +5V, for instance. There will then be some cut-off. The system will treat +/- 3V or greater as a 1 or 0. So, 3.5V is considered a 1. 4.5V is considered a 1, etc. -3.5V is considered a 0. -5V is considered a zero. This gives plenty of dead-band to help limit interference.
Next, there is some kind of language that they will speak at this hardware level. For instance, in electronics, there is often a pin that will either have a voltage applied or removed by one side or the other. When this voltage changes, one side knows that it needs to receive data, because the other informed it that it was about to send data. This is commonly called handshaking. It can also deal, at the hardware level, with data collisions on the line, where everyone waits for it to clear to send data.
Some important notes:
* I left out fiber here. It essentially works the same, but it does so with pulses of light
* It is important to remember that these networks switch VERY fast. Thousands of times a second. That is how you get your one’s and zeros for the binary transmission
* Networks have two main components: An electrical standard and a network standard. The electrical standard specifies things like cable geometry, voltage levels, power levels, pin-outs, etc so that hardware can be produced by many manufactures and won’t blow up. So that it will work together. There are lots of them. You can check some out if you want. RS232, RS485, DH+ (odd-ball early implementation of a vendor-specific electrical standard), ethernet. Those are all electrical standards that specify how the wiring is to be installed. Network standards often involve protocols, which are like languages. Examples of these are: Ethernet/IP (uses ethernet electrical standard, but is an industrial protocol), Modbus RTU (uses modbus protocol on RS485 or RS232 networks), Modbus ASCII (Like RTU but uses ASCII for data encoding), Modbus TCP (uses the ethernet electrical standard, but with modbus RTU protocol), Ethercat (uses ethernet standard, but with very high speeds), Canbus (serial bus-type electrical protocol AND standard used in some industrial drive applications and pretty much every automobile). The list goes on, and as you can see, it is a fucking mess.
Then, we move up a level. This is the protocol. The protocol is the language that they speak. Where the hardware gets the signals, it passes it up a level, where the information is read and actually processed. This is where you store that data you receive and manipulate it. Check it for errors, etc. It is important to remember that sometimes, this protocol level stuff is actually implemented in hardware for specific applications to speed it up. I’m not touch a whole lot on this, because it is very complex.
Now, in summation, the hardware is designed in such a way that it shares common voltage levels and circuitry. Then, voltage levels / light pulses “happen”, which is essentially just the circuit being switched very quickly. These events are then read, recorded, tracked and eventually stored and processed.
Other useful tidbits:
* FET-type transistors make these crazy switching speeds possible.
* OSI-model for data transfer. Covers the hardware on up to the actual application. Each layer is an abstraction on top of the other, with the lower numbers being more “basic” and fundamental.
* Remember to pay attention to install standards – they matter!
It is okay to be intimidated. This is a super-complex topic and even after years and years, no one knows all about all of it. Start by learning about basic Serial RS232 and wire it up, send data, play with it. Once you get an understanding there, you can move on to more involved stuff like ethernet and USB.
Latest Answers