Features extraction for PSK signals recognition using nonuniform compressive samples based on high order transformation

The N
^th
Power Nonlinear Transformation (NPT) is a common method for automatic modulation classification, especially for PSK signals. However, sampling rate required in NPT is much more than Nyquist rate, which is a huge burden for the Analog to Digital Converter (ADC). In this paper, we show that the N
^th
power nonlinear features used for PSK signals are compressible in the Amplitude domain. Thus, here, we propose a novel Automatic Modulation Recognition (AMR) method which based on the N
^th
power nonlinear features of PSK type signals by introducing the compressive sensing (CS) theory.