To Improve Error Protection and Security of Data the Data is Interleaved and its type is Diagonal let’s check how it works.
The diagram below illustrates, in a simplified form, diagonal interleaving applied to a 9.6 kbit/s data channel.
The diagram shows a sequence of ‘data blocks’ after the encoding process previously described, all from the same subscriber. Each block contains 456 bits, these blocks are divided into four blocks each containing 114 bits. These blocks are then interleaved together.
The first 6 bits from the first block are placed in the first burst. The first 6 bits from the second block will be placed in the second burst and so on. Each 114 bit block is spread across 19 bursts and the total 456 block will be spread across 22 bursts.
Data channels are said to have an interleaving depth of 22, although this is sometimes also referred to as an interleaving depth of 19.
The data bits are spread over a large number of bursts to ensure that the data is protected. Therefore, if a burst is lost, only a very small amount of data from one data block will actually be lost. Due to the error protection mechanisms used, the lost data can be reproduced at the receiver.
This wide interleaving depth, although providing a high resilience to error, does introduce a time delay in the transmission of the data. If data transmission is slightly delayed, it will not effect the reception quality, whereas with speech, if a delay were introduced this could be detected by the subscriber. This is why speech uses a shorter interleaving depth.