A Low Complexity and Flexible Implementation of 2n-QAM for Software Defined Radio Applications

A low complexity and flexible implementation of Gray-coded 2(n)-QAM with rectangular constellation is proposed and discussed, where it drastically saves the required storage at the cost of very low complexity. The proposed approach introduces a unified implementation algorithm for Gray-coded 2(n) QAM, for all n, where it is favorable when the radio needs to adaptively increase or decrease the modulation code-word size based on the required data rate and the channel status. The performance of a successive interference cancellation algorithm is evaluated as optimal detector using computer simulations. The proposed approach has flexibility for selectively loading the code-word's bits with unequal error rates. It is shown that for the odd values of n, stretching the constellation simultaneously reduces the peak to average power ratio (PAPR), and improves the bit error rate. An experimental implementation of the modulation on a low power ARM microcontroller is presented.