The first CH octet is mandatory and is defined as follows:

As noted in clause 4, the compression algorithms can be configured to operate in a variety of ways and may rely on end-to-end knowledge of "prior" information such as which key word dictionary is to be used. A requirement that all configuration information be explicitly stated in the CH is less efficient (in terms of compression ratio) than if a default configuration is known and only variations from this need be signalled. However, a major determinant of configuration is the language in which the original message to be compressed is composed. For example, different keyword dictionaries would be required for French and opposed to German and character frequency distributions for English texts may vary greatly from those for Swedish texts. From this it can be seen that a universal "default" configuration would be of little value. To address this, the Compression Language Context (CLC) allows a default configuration to be specified for each of the languages defined in TS 23.038 in relation to the Cell Broadcast Data Coding Scheme as follows:

• The CLC in bits 6 to 3 of the CH specify the language as per TS 23.038 in the case where bits 7 to 4 of the Cell Broadcast Data Coding Scheme octet are set to 0000.
• If and when required, higher order bits of the CLC can be signalled by a subsequent CH octet as described below.
• The CLC value 1111 (language unspecified) will indicate a "default" configuration that is language independent. This is specified in Annex R and involves the basic Huffman (de-)coding with no initial character frequency distribution, see example below.

Any second and subsequent CH octets are used to vary the configuration defaults established by the CLC. These octets all comprise a continuation bit followed by a Type, Value structure as follows:

The bits of the semi-octet CH Extension value are interpreted left to right, MSB to LSB. If the CH contains more than one octet of the same CH Extension type, the CH Extension value of a subsequent CH octet, is interpreted as being next most significant semi-octet of the composite value being signalled. For example if the CLC in CH octet 1 indicates that the default Huffman Initialization ID is 1 (decimal) and the required HI-ID is 37 (decimal), then the following octets (in the range 2 to n) would also be required in the CH.

The following values are defined for the CH Extension Type: 000

001 010 011 100 101 110 111

The four component compression algorithms (Huffman, Keywords, Character Groups and Punctuation) may all have a variety of initialization options. For each algorithm, a given set of initialization options needs to be identified for the processing of a given input stream. Initialization and operation of the algorithms depends not only on the language in which the original source text is composed but also the character set (UCS2 or otherwise) that is to be used during processing. Thus the Huffman Initialization ID (HI-ID), Keyword Dictionary ID (KD-ID), Punctuator ID (PU-ID) and Character Group ID (CG-ID) only define unique values within the context of a given character set (the default established by the CLC or subsequently amended via Change Character Set or Change UCS2 Row CH Extension types) and within the context of the language indicated by the CLC.