NOVEL 2-12QAM MODULATION FORMAT

Abstract

The 16 Quadrature Amplitude Modulation (16QAM) is the preferred modulation format for many communication and broadcasting standards. However, performance of the overall “practical” communication/ broadcasting systems employing 16QAM is far from the Shannon limit as there is a requirement for power back-up to combat non-linear distortions generated by the output of high power amplifiers. In this paper we propose moving from 16QAM to a 12QAM modulation without any change of the hardware and with only minor changes in the mapping algorithm, by eliminating the four constellation points with highest signal energy. However, to ensure that no data is lost and spectral efficiency is preserved, we propose pairing two 12QAM symbols into a 4-dimensional signal with the appropriate mapping, obtaining the desired 2-12QAM modulation. We propose the use of a rate 6/7 forward error correcting (FEC) code combined with the 2-12QAM, which maps 7 binary input digits into 2 twelve-ary symbols which select corresponding signals from the conventional 16QAM modulator. The proposed new modulation format has a spectral efficiency of 3 bits/symbol, the same spectral efficiency as the legacy 16QAM scheme when the 16QAM modulation format is coupled with a rate 3/4 FEC. The removal of the 16QAM highest energy symbols represents a peak energy gain of 2.55 dB and an average energy gain of 1,35 dB which is offset by a 0.58 dB increase in code rate, from 3/4 to 6/7. In addition, the proposed scheme has a clear advantage of employing existing 16QAM modulator/demodulator hardware and maintaining compatibility with the legacy standards, albeit with simple changes in mapping which can be achieved via a software upgrade. We show that the proposed technique allows at least 2dB power gain without sacrificing spectral efficiency or additional hardware complexity.