|
|
Data Transmission
|
|
Data communications is primarily concerned
with the movement of data from one location to another.
The term data refers to alphabetical, numerical, or
special purpose characters which, when appropriately
grouped, constitute some words or message. This data
can be taken from or delivered to various types of equipment.
It can be sent to some instrument, such as a temperature
controller. It can come from a source document, such
as an inventory or payroll or from storage medium, such
as paper or magnetic tape or punched cards. It can also
come from same register within a computer. Pulses, or
digital transmission, are used in the transmission-of
such signals because most digital computers can directly
handle them. By using communication facilities that
can handle digital data, interconnected digital computers
can directly process the data.
Data communication is rapidly expanding as more and
more agencies employ computer-based communication systems.
Few of the uses of data communications are reservation
services such as in airlines and trains, mail service,
inventory control, industrial control, computer time-sharing
etc.
According to the Information Theory, we can say that
highly efficient and virtually errorless data transmission
is possible. But the vast majority of applications,
simply do not require the cost and complexity demanded
by a near-ideal communication system. However, information
theory provides the necessary understanding of arriving
at some degree of optimum systems. |
The design of a practical data system
is based on two fundamental criteria: |
- The gross source rate (binary digits per sec), as
distinguished from the entry rate (bits per sec).
It, therefore, dictates signaling speed.
- The described transmission reliability, i.e., the
tolerated error rate. It, therefore, dictates error-probability
requirements.
|
Thus, signaling speed
and error probability are the defining
parameters of digital systems. Analogous to analog systems,
signaling speed is equivalent to bandwidth and error
probability is equivalent to signal-to-noise ratio.
Similar to analog signals, the digital signals have
a wide range of bit rates depending on the source and
application. Typical bit rates may range from 100 bits/sec
for manual keyboard devices such as tele-typewriter
to 109 bits/sec for a computer. Currently, the best
wideband systems have speed capabilities of order of
107 bits/sec, accommodating all but direct computer
to computer links. Most digital transmission, however,
is via standard telephone voice circuits with top speed
of 1,000 to 5,000 bits/sec.
In contrast to signaling speed, the issue of acceptable
error rates is rather cumbersome. For systems like defense
surveillance, error is of great concern; an error probability
of 10-5 may even be intolerable. For tele-metering data
from measuring instruments having 10 % accuracy, the
error rate may not be that stringent; a probability
of 10-3 may be unnecessary. However, error probabilities
of order of 10-3 are typical and suitable for many applications.
One fundamental difference between digital and analog
transmission must be properly noticed here: In analog
transmission, the aim is to reproduce the actual message
waveforms as close as possible. In digital transmission,
the goal is to transmit the message information as accurate
as possible. Therefore, the design of digital system
is not bound to any particular waveform and can be evolved
by variety of ways in which digital message can be represented.
In addition, since a digital signal is discrete in nature,
system refinements such as regeneration and error-control
coding can be employed. It is also possible to provide
the receiver with a reference list before hand what
the message elements should look like in the absence
of noise and other contamination. Then, given the contaminated
signal at the receiver, transmission reliability can
be further improved by a matching procedure wherein
the received waveform is compared with the possible
transmitted waveforms. Referred as coherent detection,
it provides system optimization through the use of statistical
decision theory.
This chapter discusses the fundamentals of data transmission
in the context of base-band systems; it is then followed
by description of various techniques employed in digital
systems. |
|
| Back to the
top |
|
|
|
|
|
