How is the digital representation of signals different from analog?
Traditional analog representation of signals is based on similarity
(similarity) electrical signals (changes in current and voltage)
the initial signals they present (sound pressure, temperature,
speed, etc.), as well as the similarity of the shapes of electrical signals in different
personal points along the amplification or transmission path. Electric way
which curve describes (also called transfer) the original signal,
as close as possible to the waveform of this signal.
This representation is the most accurate, yet the slightest distortion of the shape
carrying an electrical signal will inevitably involve the same
distortion of the shape and signal carried. In terms of information theory,
the amount of information in the carrier signal is exactly equal to the amount
information in the original signal, and the electrical representation does not contain
there is redundancy which could protect the carried signal from being
storage, transmission and amplification.
The digital representation of electrical signals is designed to add
redundancy that protects against the effects of parasitic interference. For this-
Serious restrictions are placed on the carrier electrical signal.
– its amplitude can take only two limit values - 0 and
1. The entire area of possible amplitudes in this case is divided into three areas: the lower
ny represents zero values, the upper represents ones and the intermediate
naya is prohibited, only interference can enter.
Therefore, any interference whose amplitude is less than half the amplitude
where the carrier signal is, it does not affect the correct transmission
values 0 and 1. Interference with a higher amplitude also does not affect
if the duration of the interference pulse is noticeably shorter than the duration
information pulse, and a filter is installed at the input of the receiver.
pulse interference.
The digital signal formed in this way can carry any
useful information encoded as a sequence
bits: zeros and ones; special cases of such information are
electrical and sound signals. Here, the amount of information carried
digital signal is much more than encoded output
nom, so that the carrier signal has a certain redundancy with respect to
original, and any distortion of the waveform of the carrier signal, when
which still retains the receiver’s ability to correctly distinguish zeros
and the units do not affect the reliability of the information transmitted by this signal.
formations. However, in the case of significant interference, the waveform
may be distorted so that the accurate transmission of the information carried
becomes impossible: errors appear in it, which, with a simple
Encoding method, the receiver can not only correct, but also detect
to shoot.
To further enhance the immunity of the digital signal to interference and usage
It appears that redundant digital encoding of two types is used:
correct (EDC – Error Detection Code) and
correct (ECC – Error Correction Code)
codes. Digital encoding consists of a simple addition to the original.
formation of additional bits and / or transformation of the original bit
chains in a chain of greater length and different structure. EDC allows
just to detect the fact of the error – distortion or loss of utility or
the appearance of a false digit, however, the information transferred in this case
also distorted; ECC allows you to immediately correct detected errors –
ki, keeping the information that is transported unchanged. For convenience and reliability
the transmitted information is divided into blocks (frames), each of which
supplied with its own set of these codes.
Each type of EDC / ECC has its own capacity limit to detect and
correct errors, followed by undetected errors again and
distortion of the information that is transferred. The increase in EDC / ECC volume is related to
the amount of initial information in the general case increases detection
the ability to correct and correct these codes.
Like EDC, the popular cyclic redundancy code CRC (Cyclic
Redundancy check), the essence of which is the complex mix of
input information into the block and the formation of short binary words,
whose ranks are in strong cross-dependence on each
block bit. Changing even one bit in a block causes significant
the change in CRC calculated from it, and the probability of such a bi-
what the CRC does not change is extremely small, even for a short time
(units of percentage of the length of the block) CRC words. As an ECC, use
Hamming and Reed-Solomon codes are given, which
it also includes EDC functions.
