as having two ports: one for transmitting and one for receiving. The ports can be thought of as consisting of a single wire, or at most two wires. These simplifications are possible and reasonably accurate.
There are two signalling methods for data transmission among communications links: asynchronous and synchronous. The application (type of data transmitted and data rate) and history determine which of the methods are used. The following description of these two signalling methods is based on Murphy and Kallis (1968).
In asychronous serial data transmission, characters (encoded into streams of bits) are transmitted one at a time, whenever the line is not busy. A continuous stream of characters can be sent, end to end, but this is not necessary. The presence of each character is indicated by special start/stop codes, which appear on the line. Asynchronous transmission has the following advantages:
a. Data is easily generated and detected by electromechanical equipment (e.g. Teletype keyboards and printers).
b. Characters can be sent at an "asynchronous" rate (i.e. at will, as long as the line is not busy), because each character carries its own synchronizing information.
The disadvantages of this technique are:
a. It is distortion sensitive. The receiver depends upon incoming signal sequences becoming synchronized. Any distortion in these sequences will affect the reliability with which the character is assembled; hence characters are usually limited to eight bits.
b. It is speed limited. To accommodate distortion, transmission speeds can only go as high as about 2000 baud (bits/sec).
c. It is inefficient. At least 10 bit-times are required to send 8 data bits. If a 2-bit start/stop code is used, 11 bit times are required to transmit 8 data bits.
In the synchronous serial technique, a continuous stream of characters is sent over the line, and there is usually no need for special start/stop codes for each character. Instead, characters are separated by transmitting a unique code at the beginning of a character stream which, when recognized, causes the receiver to lock in (frame) the incoming bits and assemble them as characters. As in the asynchronous technique, the character length (in bits) is fixed. Other full character codes signify other conditions concerning the data stream.
Unlike the asynchronous technique, a synchronizing signal must be provided along with the data bit stream. This signal is in the form of a clock which signifies when each data bit can be transmitted or has been received; bits are transmitted only at clock times, and at each clock time, either a 0 or a 1 must be transmitted. This clock signal can either be provided by the transmitter, or by some separate source that the transmitter uses for timing. The format for this type of transmission is shown in figure TRAN-2.
In the format shown, the transmitter presents data to the line on the negative going (+ to -) transition of the timing signal, and the receiver samples the data line on the timing signal positive transition (- to +).
The advantages of the synchronous serial techniques are:
a. A common timing source can be used for both transmitter and receiver, hence clock-synchronizing logic is minimal.
b. Efficiency is increased, since there are no bit-times wasted with the use of start/stop code bits. All bits on the line are data, with the exception of a single synchronizing pattern at the beginning of the bit stream.
c. There is low distortion sensitivity, due to the timing being provided with the data, allowing higher speeds.
232