Unitary space-time modulation identification based on high order statistics

Modulation identification for communication signals is an important topic in the intercepted signal processing. However, early researches mainly focused on common modulation. Aiming at the statistical feature of unitary space-time modulation (USTM), we propose the suitable algorithms for high order joint moment and high order joint cumulant, and employ the differences of three order statistics to separate UST signals from MPSK signals and Gaussian noise, so we can complete the modulation identification of USTM. Simulation results show that the feature parameters can successfully suppress the Gaussian noise. The proposed algorithm is simple to realize and has a high probability of correct identification even in the low SNR region. It can also perform well in fading channels.