An Introduction to Adaptive QAM Modulation Schemes for Known and Predicted Channels

A major disadvantage with fixed modulation (nonadaptive) on channels with varying signal-to-noise ratio (SNR) is that the bit-error-rate (BER) probability performance is changing with the channel quality. Most applications require a certain maximum BER and there is normally no reason for providing a smaller BER than required. An adaptive modulation scheme, on the contrary, can be designed to have a BER which is constant for all channel SNRs. The spectral efficiency of the fixed modulation is constant, while it, in general, will increase with increasing channel SNRs for the adaptive scheme. This in effect means that the average spectral efficiency of the adaptive scheme is improved, while at the same time the BER is better suited to the requirement of the application. Thus, the adaptive link becomes much more efficient for data transmission. The major disadvantage is that the transmitter needs to know the channel SNR such that the best suitable modulation is chosen and the receiver must be informed on the used modulation in order to decode the information. This leads to an increased overhead in the system as compared with a fixed modulation system. In this paper, we introduce adaptive modulation systems by presenting some of the simpler adaptive quadrature amplitude modulation schemes and their performance for both perfectly known and predicted channels.