Atsc working Draft Template

Compression
19
 (Normative) 
C.1 SCOPE 
This Annex describes the compression method adopted for the transmission of English-language 
text strings in PSIP. The method distinguishes two types of text strings: titles and program 
descriptions. For each of these types, Huffman tables are defined based on 1st-order conditional 
probabilities. Section C.3 defines standard Huffman encode and decode tables optimized for 
English-language text such as that typically found in program titles. Section C.4 defines Huffman 
encode and decode tables optimized for English-language text such as that typically found in 
program descriptions. Receivers supporting the English language are expected to support decoding 
of text using either of these two standard Huffman compression tables. 
The encode tables provide necessary and sufficient information to build the Huffman trees that 
need to be implemented for decoding. The decode tables described in Tables C5 and C7 are a 
particular mapping of those trees into a numerical array suitable for storage. This array can be 
easily implemented and used with the decoding algorithm. However, the user is free to design its 
own decoding tables as long as they follow the Huffman trees and rules defined in this Annex. 
C.2 CHARACTER SET DEFINITION 
This compression method supports the full ISO/IEC 8859-1 [11] (Latin-1) character set, although 
only characters in the ASCII range (character codes 1 to 127) can be compressed. The following 
characters have special definitions: 
Table C1 Characters with Special Definitions 
Character 
Value 
(Decimal) 
Meaning 
String 
Terminate 
(ASCII Null) 
0 
The Terminate character is used to terminate strings. The Terminate character is 
appended to the string in either compressed or uncompressed form. 
The first encoded character in a compressed string is encoded/decoded from the 
Terminate sub-tree. In other words, when encoding or decoding the first character in a 
compressed string, assume that the previous character was a Terminate character. 
Order-1 
Escape 
(ASCII 
ESC) 
27 
Used to escape from first-order context to uncompressed context. The character which 
follows the Escape character is uncompressed.