CCS - Analog-to-Digital Conversion

 

Analog-to-Digital Conversion

The majority of information sources are analog by nature. Analog sources include
speech, image, and many telemetry sources. There are various methods
and techniques used for converting analog sources to digital sequences in an efficient
way. This is desirable because digital information is easier to process, to communicate, and to store.

The general theme of data compression, of which analog-to-digital conversion is a special case, can be
divided into two main branches:

 

Quantization (or lossy data compression), in which the analog source is quantized
into a finite number of levels. In this process some distortion will inevitably
occur, so some information will be lost. This lost information cannot be recovered.
General analog-to-digital conversion techniques such as pulse-code modulation
(PCM), differential pulse-code modulation (DPCM), delta modulation
(􀀞M), uniform quantization, nonuniform quantization, and vector quantization
belong to this class. The fundamental limit on the performance of this class of
data-compression schemes is given by the rate-distortion bound.

 

Noiseless coding (or lossless data compression), in which the digital data (usually
the result of quantization, as discussed above) are compressed with the goal
of representing them with as few bits as possible, such that the original data
sequence can be completely recovered from the compressed sequence. Sourcecoding
techniques such as Huffman coding, Lempel-Ziv coding, and arithmetic
coding belong to this class of data-compression schemes. In this class of coding
schemes no loss of information occurs. The fundamental limit on the compression
achieved by this class is given by the entropy of the source.

 

Reference,

　　1. <<Contemporary Communication System using MATLAB>> - John G. Proakis