A Fairness–Throughput Tradeoff Perspective on NOMA Multiresolution Broadcasting

We propose an analytical framework for characterizing the tradeoff between fairness and throughput that arises in a cellular or satellite downlink when multicast information is encoded in two resolution levels (high versus low priority information). Given a target fairness (measured in terms of the ratio between the information rates of low versus high resolution stream), the operator seeks to optimize the modulation and coding schemes so as to maximize the average cell throughput. Viewing operating points as fairness-throughput pairs allows to meaningfully quantify and interpret the superiority of non-orthogonal multiple access techniques such as superposition coding against simple orthogonal time division. The optimal fairness-throughput tradeoff curve for superposition coding versus time division is derived in the Gaussian setting. A practical implementation with 16-QAM constellations and multilevel coding is proposed and its tradeoff curve is numerically evaluated.