Blind Recognition and Detection for Hybrid Modulation in Grant-Free Multiple Access

Hybrid modulation can improve transmission efficiency, lower bit error rate, and increase the system’s universality and flexibility. However, traditional hybrid modulation recognition methods, such as those based on instantaneous features, are not suitable for autonomous grant-free systems. To address this issue, this paper presents a blind recognition and detection algorithm for hybrid modulation in non-orthogonal multiple access systems with autonomous grant-free. Since low-order modulation constellations have obvious geometric distributions, the number of user signal clusters can be used to identify the modulation scheme. We calculate the ideal threshold based on noise and interference modeling and propose a geometric feature-based method based on the constellation for clustering judgment to realize hybrid modulation recognition and detection. Simulation results are provided, which demonstrate the advantages of the proposed scheme in blind recognition for hybrid modulation under the same conditions.