Digital modulation recognition based on discriminative restricted Boltzmann machine

Abstract

In order to improve the performance of digital modulation recognition under high dynamic signal-to-noise ratio, a joint modulation recognition method based on high-order cumulant and discriminative restricted Boltzmann machine was proposed, which extracted the high-order cumulant of digital signals as signal features, comprehensively utilized the generation ability and classification ability of the discriminative restricted Boltzmann machine, analyzed the recognition rate of digital signals in environments containing Gaussian noise, time-varying phase offset or Rayleigh fading.Experimental results show that compared with traditional classification methods, the recognition performance of the proposed method is obviously improved.In addition, the use of the model’s generation ability to reconstruct the input features can effectively improve the signal recognition rate under low signal-to-noise ratio.

Keywords

• modulation recognition;restricted Boltzmann machine;high-order cumulant;data reconstruction