I jointly teach the module entitled “Advanced Modulation and Coding” with Dr Martin Johnston. This module is offered to both our MSc and MEng students. Dr Johnston is in charge of the “coding” part, whereas I teach the “modulation” part. You can download the lecture notes for my part by clicking on this link: EEE8003.

The coursework for this module consists of a Matlab assignment in which students have to assess via Monte Carlo computer simulations the error performance of several coded modulation systems over both Gaussian and flat Rayleigh fading channels. The assignment questions can be found in the last pages of the lecture notes.

In the Matlab assignment, students are required at some stage to use a soft-decision Viterbi decoder in order to improve the error performance of the system. This can be done only if a de-mapping function (also known as a “demapper”) is employed at the channel output. The purpose of the demapper is to convert the channel estimates into estimates of the transmitted bits that can then be processed by the soft-decision Viterbi decoder.

Finding the equations to implement the de-mapping function is not an easy task. This is why I am happy to offer a bit of help here: these equations can be found in the two following papers: ICC 1994 and Eusipco 2000.

You will notice that, remarkably, the demapper can be implemented using a set of very simple equations.