Analysis of Performance of MAC Based Iterative MUD Using EXIT and BER Charts

Iterative multi user detection (MUD) enables optimal data detection and decoding in communication channels by interacting between multiuser detection and channel decoding [11-13]. Increasing number of active users in the multiple-access channel increases the interference. To minimize the degradation, the number of iterations is usually increased at the cost of additional latency. In this paper it is shown that a slight modification in the implementation of message authentication codes (MAC) in the iterative MUD leads to a decrease in the number of iterations required for optimal detection and also improves the performance of the multiuser receiver.

In this paper the iterative MUD has the MAC bits spread out in the convolutional coded data block. These uncoded MAC bits are used as apriori data bits in the iterative decoding. This results in a reduction of the number of iterations required for the same BER. An EXIT chart of the combination of the convolutional decoder block with that of the multi user detector is created and the BER results are validated.