Transmitting digital information

[Jimmy87]

Hi pf, please could someone explain in detail how binary information is physically transmitted. For example, consider an analogue to digital converter which is sampling a sine wave representing a voice. The sampling is using 8 bits. At a certain time let's say the sine wave is at its peak voltage of 5 volts. This means that the binary code for this peak would be the 255th value with a binary value of 1111 1111. My question is how is this 1111 1111 physically transmitted in a wire? Because these are just numbers. Would it simply be an electrical voltage pulse of 5 volts which itself is made up of a series of eight 1's in this case created by individual transistors?

 

[D H]

There are a number of different techniques for transmitting (and receiving) digital data. A good place to start is the wikipedia article on modulation.

There's always a carrier signal that is somehow modulated in some way to represent that data that is to be transmitted. This modulation can involve changing the amplitude of the carrier signal, changing the frequency of the carrier signal, changing the phase of the carrier signal, or some combination of these techniques.

 

[M Quack]

That depends very much on the protocol used.

In a parallel interface, there is one wire for each bit. '0" could be 0V, and "1" could be 5V. Additional wires could be used for synchronization. This type of interface was used on the old centronics-type parallel printer ports of computers

http://en.wikipedia.org/wiki/Printer_port

Serial interfaces use a much smaller number of wires. All the data travels through the same wire, one bit after the others. A textbook example is the RS232 "serial" port

http://en.wikipedia.org/wiki/Rs232

There are about as many interfaces and protocols as there is sand on the sea shore... I2C, USB, SATA, SCSI, GPIB, S/PDIF, Ethernet... In all cases, the electrical specifications go hand in hand with the timing, synchronization, data type, etc specifications.

 

[Jimmy87]

That depends very much on the protocol used.

In a parallel interface, there is one wire for each bit. '0" could be 0V, and "1" could be 5V. Additional wires could be used for synchronization. This type of interface was used on the old centronics-type parallel printer ports of computers

http://en.wikipedia.org/wiki/Printer_port

Serial interfaces use a much smaller number of wires. All the data travels through the same wire, one bit after the others. A textbook example is the RS232 "serial" port

http://en.wikipedia.org/wiki/Rs232

There are about as many interfaces and protocols as there is sand on the sea shore... I2C, USB, SATA, SCSI, GPIB, S/PDIF, Ethernet... In all cases, the electrical specifications go hand in hand with the timing, synchronization, data type, etc specifications.

Thanks for the answers guys. Just been trying to digest the info you both gave me. Say if you sample the same sine what I was talking about a fraction of a second later and it was 3.2V and that this corresponded to a binary code of 1101 0110 ( I just made this up). How would you transmit 1101 0110 back to a digital to analogue converter for example. Would you send electrical pulses to transmit each 1 and 0 or is it modulated onto a carrier wave? I have done modulation before on radio transmission but didn't know it applied to a computer as well. So is there a carrier wave inside an ADC found inside a computer?

 

[skeptic2]

The simplest would be to transmit a carrier for a 1 and no carrier for a 0. This doesn't work well if you want to send a string of 1s or a string of 0s. There are ways of adding a clock onto the data in order to decode these types of strings. One such is Manchester encoding. However Manchester encoding reduces the data bandwidth by a factor of 2.

With FM, you could use + & - deviation for 1s and 0s. By adding more levels you can transmit more than one bit per symbol. C4FM is a common method of sending data with FM which uses 4 deviation levels to send 2 bits at a time.

Increasing in sophistication there is QPSK and its many derivatives which use multiple AM and FM levels to send multiple bits per symbol.

Two more ways of sending multiple bits per symbol are Pulse Width Modulation and Pulse Position Modulation.

In your example above, wouldn't it be more efficient to just transmit the difference between the last value and the current value. Instead of sending the whole value. Instead of sending 11010110 you would only need to send -101001. A type of modulation that does that is called Delta Sigma modulation.

 

[Jimmy87]

The simplest would be to transmit a carrier for a 1 and no carrier for a 0. This doesn't work well if you want to send a string of 1s or a string of 0s. There are ways of adding a clock onto the data in order to decode these types of strings. One such is Manchester encoding. However Manchester encoding reduces the data bandwidth by a factor of 2.

With FM, you could use + & - deviation for 1s and 0s. By adding more levels you can transmit more than one bit per symbol. C4FM is a common method of sending data with FM which uses 4 deviation levels to send 2 bits at a time.

Increasing in sophistication there is QPSK and its many derivatives which use multiple AM and FM levels to send multiple bits per symbol.

Two more ways of sending multiple bits per symbol are Pulse Width Modulation and Pulse Position Modulation.

In your example above, wouldn't it be more efficient to just transmit the difference between the last value and the current value. Instead of sending the whole value. Instead of sending 11010110 you would only need to send -101001. A type of modulation that does that is called Delta Sigma modulation.

Thanks for your answer. Could you expand on how you would send multiple bits. For example, consider a telephone modem which changes your voice into a digital signal by using different tones - a certain tone for a 1 and a different tone for a 0. Say the modem was to send the letter 'a' which has an ASCII code of 01100001. How would a modem send a string like this or would it just send one bit at a time?

 

[skeptic2]

This example is by no means the most efficient. In an FM system let's define 4 deviation levels each representing 2 bits, with 1 equal to maximum deviation. The 4 levels would be -1 = 00, -.333 = 01, +.333 = 10 and +1 = 11. The bit combinations can easily be converted to analog levels and those fed into the modulator to get the four levels.

At the receive end the discriminator will give you 4 analog levels out which can easily be converted back to the bit pair.

 

[analogdesign]

Thanks for your answer. Could you expand on how you would send multiple bits. For example, consider a telephone modem which changes your voice into a digital signal by using different tones - a certain tone for a 1 and a different tone for a 0. Say the modem was to send the letter 'a' which has an ASCII code of 01100001. How would a modem send a string like this or would it just send one bit at a time?

Telephone modems are not a good choice to understand digital communications because they use remarkably complex modulation schemes. Keep it simple. To answer your question, it could send one bit at a time (this would be the simplest) but in practice it would send multiple bits at a time by using the phase relationship between two tones or by using different tones. This is too complex for you to start with though.

The fact is there are many, many ways you can convey digital information. For example you could use morse code. Or as skeptic2 says you could define presence of carrier = 1 and absence of carrier = 0 (as above). This is a very simple modulation scheme. Even simpler would be to use three wires. On one wire put a clock. Then on another wire put 5V (say) for 1 and 0V (relative to the third, ground wire) for zero. The receiver looks at the voltage on wire #2 every time it senses a rising edge of the clock. This would make a simple and reliable communications channel.

It would be frightfully inefficient, of course. Communications schemes in practice use various techniques to get rid of the clock wire and are amazingly efficient. Believe it or not, we can even prove the maximum datarate possible based on the characteristics of a communication channel. This is called the Shannon Limit. I suggested you avoid telephone modems because they are extremely close to the Shannon Limit for a telephone wire.

 

[marioxcc]

Jimmy87, as already said, there are several ways to an analog signal digital information as an analog signal. It don't makes sense that you ask how a modem would transmit a given binary sequence because it depends on the modulation method that it uses. If you want to know which modulation method is used by a specific device, then you must look into the protocol it uses.

Modulation methods can be baseband or passband. Baseband methods generate analog signals without necessarily having a bound of the lower frequency present. Passband generate signals with frequencies within a prescribed range. See the case of RS-232 and 8b/10b for a baseband examples and FSK and QAM for passband examples.
---
This message is Copyright © 2014 Mario Castelán Castro and it is licensed under the Creative Commons Attribution-ShareAlike 4.0 International license.

 

[skeptic2]

consider a telephone modem which changes your voice into a digital signal by using different tones - a certain tone for a 1 and a different tone for a 0. Say the modem was to send the letter 'a' which has an ASCII code of 01100001. How would a modem send a string like this or would it just send one bit at a time?

From Wikipedia: http://en.wikipedia.org/wiki/Modem

The famous Bell 103A dataset standard was also introduced by AT&T in 1962. It provided full-duplex service at 300 bit/s over normal phone lines. Frequency-shift keying was used, with the call originator transmitting at 1,070 or 1,270 Hz and the answering modem transmitting at 2,025 or 2,225 Hz. The readily available 103A2 gave an important boost to the use of remote low-speed terminals such as the Teletype Model 33 ASR and KSR, and the IBM 2741. AT&T reduced modem costs by introducing the originate-only 113D and the answer-only 113B/C modems.

The 300 bit/s modems used audio frequency-shift keying to send data. In this system the stream of 1s and 0s in computer data is translated into sounds which can be easily sent on the phone lines. In the Bell 103 system, the originating modem sends 0s by playing a 1,070 Hz tone, and 1s at 1,270 Hz, with the answering modem transmitting its 0s on 2,025 Hz and 1s on 2,225 Hz. These frequencies were chosen carefully; they are in the range that suffers minimum distortion on the phone system and are not harmonics of each other.

In the 1,200 bit/s and faster systems, phase-shift keying was used. In this system the two tones for anyone side of the connection are sent at similar frequencies as in the 300 bit/s systems, but slightly out of phase. Voiceband modems generally remained at 300 and 1,200 bit/s (V.21 and V.22) into the mid-1980s. A V.22bis 2,400-bit/s system similar in concept to the 1,200-bit/s Bell 212 signaling was introduced in the U.S., and a slightly different one in Europe. The limited available frequency range meant the symbol rate of 1,200 bit/s modems was still only 600 baud (symbols per second). The bit rate increases were achieved by defining 4 or 8 distinct symbols, which allowed the encoding of 2 or 3 bits per symbol instead of only 1. The use of smaller shifts had the drawback of making each symbol more vulnerable to interference, but improvements in phone line quality at the same time helped compensate for this. By the late 1980s, most modems could support all of these standards and 2,400-bit/s operation was becoming common.

 

[Devanand]

Hi Jimmy, In Computer system everything you transmit or receive will be in digital form there is no need of Radio techniques used within this system. If u sample '5' volt using ADC the output will be '11111111'(according to your statement), This '8 bit' digital data will be stored in a Register(ADC) before it is being sent to any other peripheral for eg: DAC.

The communication of this Data within the computer system is in the form 'Digital Pulses'( 1 and 0 ), where '1' can be any voltage which can be recognized by that particular system and '0' normally will be at ground level.

You don't need to follow any protocol to send and receive data unless the Receiver and the Transmitter follows any (Speaking of only digital System).

 

[analogdesign]

Hi Jimmy, In Computer system everything you transmit or receive will be in digital form there is no need of Radio techniques used within this system. If u sample '5' volt using ADC the output will be '11111111'(according to your statement), This '8 bit' digital data will be stored in a Register(ADC) before it is being sent to any other peripheral for eg: DAC.

The communication of this Data within the computer system is in the form 'Digital Pulses'( 1 and 0 ), where '1' can be any voltage which can be recognized by that particular system and '0' normally will be at ground level.

You don't need to follow any protocol to send and receive data unless the Receiver and the Transmitter follows any (Speaking of only digital System).

This is a good answer and I realize the OP was asking a very simple question.

To the OP: basically the output to your analog-to-digital converter would be 8 different wires, one for each bit. If the bit is '0', the voltage on the corresponding wire is 0V. If the bit is '1', the voltage on the corresponding wire is 5V (in your example). Simple as that.

 

[sophiecentaur]

Hi Jimmy, In Computer system everything you transmit or receive will be in digital form there is no need of Radio techniques used within this system. If u sample '5' volt using ADC the output will be '11111111'(according to your statement), This '8 bit' digital data will be stored in a Register(ADC) before it is being sent to any other peripheral for eg: DAC.

The communication of this Data within the computer system is in the form 'Digital Pulses'( 1 and 0 ), where '1' can be any voltage which can be recognized by that particular system and '0' normally will be at ground level.

You don't need to follow any protocol to send and receive data unless the Receiver and the Transmitter follows any (Speaking of only digital System).

I think you mean that there is no use of Modulation in many computer interfaces. If you use WiFi, of course, the modulation of a radio frequency carrier is involved.

Many people fail to realize that, in any digital system, the actual digital information is, in fact, carried in analogue form (i.e. in the varying voltage on (or current through) a wire. It will very seldom look like the nice rectangular 'on/off/off/off/on/on/off/on' that is drawn in elementary texts because such a system would be running at a much lower data rate than it's capable of. Throughout history, the components have been pushed to work as fast as they can and the individual digits, seen on an oscilloscope, will look just like a raggedy pattern of waves. To get the digital content of this analogue waveform, it is often necessary to use some very sophisticated analogue methods, which filter and sample the waveform at the right times and 'decide' whether a 1 or 0 has been sent at a particular time. You can't 'do' serious digital without also 'doing' analogue.

This link has some pictures of the sort of thing I mean. It shows how symbols can interfere with each other when the bandwidth is limited.