How do cable lines on telephone poles transmit and receive data along thousands of houses and not get interference?
In: Engineering
I’m not crazy about most of the responses. Telephone lines use a pair of wires. The signal is the difference between the pair. Because they are the same length, connected the same way, and right next to each other, they tend to receive the same interference–but, this doesn’t affect the difference between the pairs. Wire pairs are actually twisted to ensure that on average, each is the same distance from interferers , such as electric lines on the same pole.
Cable TV uses coaxial cables. That’s a different kettle of fish, but it’s also constructed to mitigate interference effects.
It should also be mentioned that you DO get interference. Lots of it! Most you don’t even notice, because the compounded error correction between the various pieces of hardware and software make it seem like everything is flowing smoothly.
In analog voice communications particularly, the required quality of the physical cabling (in some cases, tin or even lead) was incredibly tolerant. Many of us remember background humming, a distant-sounding busy signal, or just crackly noise during long-distance calls. Our brains did all the error correction, or we just asked the person on the other end to repeat it.
When we started using phone lines for digital communications, trying to push more than about 10Kb on the wire was unreliable. That’s when higher speeds were only realized by improving error correction. Since that time, vastly improved media (twisted copper and fiber optic are the most common now) have helped a lot, but transmission and receive errors are always present, even in communications between a processor and a hard drive.
Cable and telephone protect against interference in different ways. Telephone lines use twisted wire. The twists of the wire pairs block outside interference and help propogate a signal further. Than if it were untwisted.
Cable/coax cables run on a single copper conductor, surrounded by a dialectic medium and protected by aluminum shielding.
Both are limited by a rate/reach issue… The higher the frequency/data rate, the lower the distance. 128 kbps (very slow) dsl can go about 20 thousand feet, while a 100 Mbps dsl connection goes about 1000 ft max. Each home has an individual coax wire or twisted pair ran to it.
The cable/telephone company runs fiber to distribution locations and uses equipment (DSLAM in the case of DSL) to put the signal onto the copper wire to make the last mile connection… So in the case of them suddenly offering higher speed, it means they’ve ran a new DSLAM or fiber node closer to your home.
If you are referring to how over a hundred people can transmit data over a few wires without getting mixed up, here is the closest explanation i can muster.
Whenever you wish to send or receive data, whether it be streaming a video or downloading a file, your PC will communicate will make a request for the data. In the case of streaming a video, your device will send a request to the server holding that video. The server will then package that data by splitting it up, associating your device’s IP address (like a home address but for computers), and will pack the segments into things called packets. The packets are transmitted through the network on a wire or by radio waves in the form of binary. Devices called routers will take this packet, read the destination IP address associated to this packet and will route it accordingly. This means the wire is filled with little segments of addressed data traveling based on their destination address, much like physical mail or cars on a highway. You should not receive anybody else’s data as your IP address and/or default gateway (your router’s IP address) is always unique. But that’s my best explanation without getting into more complicated things like the OSI reference model, TCP/UDP protocols, and ports.
Some of the answers here are interesting but don’t actually address the question.
The answer is that data transmitted along cable lines DOES get interference, BUT several important things happen:
1. Information being carried down a line is some kind of analog signal. It might use various voltages and frequencies, and splice many “data streams” together in a process called multiplexing, but it’s an analog signal of some sort. However, what it represents is usually digital data, and as long as you can uncover that original digital representation okay at the other end, the interference is irrelevant.
2. Interference is okay, so long as the Signal to Noise ratio is high, meaning we can clearly distinguish some light noise from a strong signal, and ignore it. If we have to do this over really long distances maybe we put repeaters every once in a while. This way the interference over the whole distance doesn’t sum up: each repeater can strip out mild noise then repeat a clean signal down the line. Also, if we expect interference at e.g. 2MHz because e.g. it’s known to be widely used for something, maybe we transmit at 3MHz to avoid it.
3. At some point we convert back to digital. Let’s just make up a simple scenario that the signal is represented by voltage and 5v is a binary 1 and 0v is a binary 0. Well if we have about 1v of noise from interferenece and we can still easily say “anything from 0-2v is a binary 0 and anything from 3-5v is a binary 1”, then that 1v of noise doesn’t matter and the digital data we can recover will be the exact same one we put in.
4. Sometimes errors happen. Usually higher-level protocols can detect them. Say we send 011010 but we receive 011000. Well we might want a way to know something went wrong. A really simple example is called parity, and so e.g. we can just make a rule that for every 7 bits of data we add an extra bit to make the number of 1’s an even number. So we send 0110101, and we receive 0110001. We know there was an error because there are three 1’s, and that’s not an even number. The communcations protocols handling the data exchange can send a signal down the wire asking for that data to be sent again.
So the answer is that there is interference, you try to keep it low relative to the signal, you might engineer your signal characteristics to avoid certain expected interference, you might use repeaters, you represent data in a way that is less affected by interference, and you detect interferenece and handle error correction and retransmission as needed.
Latest Answers