Computers and Serial Communication

Back to the previous section, The Electric Telegraph

2. Serial Communication and Radio

Radio waves hold a unique place among the physical phenomena that are exploited by all the gadgets used in the modern world. Their existence was predicted, on a purely theoretical basis, by James Clerk Maxwell around 1873. At that time nobody knew how to make a radio wave, but a long line of dedicated experimenters slowly developed the technology of radio. Early radio equipment could generate and detect little more than clicks, so once again letters had to be built up from a serial sequential code so that communication was possible. Radio pioneers borrowed from the technology of the telegraph, and by the time of Guglielmo Marconi's famous experiment in 1901, Morse code was established as the de-facto standard for radio communication. In that experiment, the letter 's' was transmitted across the Atlantic ocean, in Morse code.

While radio transmission of Morse code eliminated the cost of telegraph wires, it did not eliminate the need for trained people to send and receive the messages. The invention of the vacuum tube triode in 1907 made voice radio broadcasting practical, but the ability to communicate over longer distance with smaller transmitting power kept Morse code in use for a long time.

Teletypewriters had been developed to eliminate the need for trained Morse code operators, and the combination of radio communication with teletypes is called radioteletype, sometimes abbreviated RTTY. Vacuum tube transmitters were able to send out continuous radio waves, and for Morse code the transmitter was simply switched on and off by the telegraph key. Special circuitry at the receiving end would turn the radio signal into a tone, so Morse code, which had previously been interpreted as clicks, became a series of long and short beeps.

Turning the transmitter on and off for radioteletype did not work as well, however, because radio noise when the transmitter was off could trigger the teletype and a letter would be typed when none had been transmitted. For a human Morse code operator, the noise was easily ignored. To improve the reliability of radioteletype, engineers used a different method: the radio transmitter was left on continuously, but the frequency was changed by the teletype signal. Two closely spaced frequencies were used, one representing the mark condition and one the space condition. In the radio receiver, the same method used to detect Morse code produced two different audio tones, and circuits were designed to detect them and recreate the on-off switched signal, which was used to drive a teletype. This type of radio modulation is called Frequency Shift Keying, and was the basis for later use of telephone lines for this type of communication.

Radioteletype was not wildly successful, since Morse code still worked better at long distances and under noisy radio conditions. Where RTTY worked well, voice transmission also worked, but RTTY had some unique advantages: messages were automatically recorded, even when the operator was not there, there was never any confusion about spelling or punctuation, and messages could be encrypted so that they could be kept secret. For these reasons, radioteletype was widely used by news organizations and the military, and networks of teletypes, both radio and wired, replaced Morse code for Telegram type communication services.

In the United States, the 1920s saw an explosive growth in the use of telephones, and telephone wires replaced telegraph wires all over the country. But telegraph lines required a continuous flow of current, and telephone exchanges transferred the voice signal from one line to another through transformers, which blocked the flow of direct current. So telephone lines were unsuitable for direct use by telegraph equipment, and they were unsuitable for teletypes as well, since they used the same direct-current signalling method.

But we have seen how radio equipment turned the clicks of the early telegraph into beeps, and those beeps were easily transmitted over telephone lines, as was the audible frequency shift keyed tone of a received radioteletype signal. There are enough possible frequencies on a telephone line that several low speed telegraph or teletype signals can be sent on a single voice channel by using different frequencies for each. This is known as frequency division multiplexing, and is similar to the methods tried by Alexander Graham Bell to increase the capacity of telegraph lines.

This is how the first modems came about. One audible tone represents the marking condition, and another the spacing condition. The tones are created by an oscillator whose frequency can be changed by the teletype signal. At the receiving end special circuits detect each of the frequencies used, and reconstruct the original signal. Since direct current is not used, the signals may be sent over telephone lines.

On to the next section, Early Computer Interfaces
Back to the Index
feedback to the author