EDGE MCS Coding Schemes
- a summary, overview or tutorial about the basics of the GSM EDGE data coding and error correction and MCS classes.
Most of the data being sent over an EDGE link will consist of TCP/IP packets. These packets are longer than a single EDGE packet payload and therefore it is necessary to split the TCP/IP packets into smaller section and these are known as "chunks". These chunks have defined sizes and may consist of one of 22, 28, 34, or 37 bytes or "octets". The 37 octet chunk may be made directly of data to be transmitted, or it may be a 34 octet chunk which is then padded by adding three dummy octets.
There are nine different Modulation and Coding Schemes (MCS) that can be used with EDGE. Each one is designated a number in the region 1 to 9. These allow different degrees of error protection (and coding rate) and this results in a change in the net data throughput. The system detects the number of bit errors and adjusts the coding scheme accordingly. It naturally endeavours to adopt the scheme that will result in the highest throughput, but will adjust itself according to the prevailing conditions, changing as required.
The different coding schemes are grouped into three classes or families which are referred to by letters, as classes A, B and C. The coding schemes within a class are used together and complement each other. Family A consists of MCS-3, MCS-6, MCS-8, and MCS-9. Family B consists of MCS-2, MCS-5, and MCS-7. Finally family C consists of MCS-1, and MCS- 4. The advantage of grouping the families together in this way is that if a block transmitted in one of the coding schemes is not acknowledged, then it can be sent as two blocks, for example with a coding scheme in the same family. For example if a block transmitted using MCS-7 is corrupted then it can be re-sent as two blocks using MCS-5 or four using MCS-2.
|MCS Scheme Name||Effective Coding rate||Modulation Format||Data Rate for One Slot
By Ian Poole
Share this page
Want more like this? Register for our newsletter