Clustering based distribution fitting algorithm for Automatic Modulation Recognition

Automatic modulation recognition (AMR) is an expert in modulation type identification. Many existing algorithms attempt to recognize the modulation candidates using phase and magnitude feature extraction. Performance is a major drawback of this feature extraction under noisy environment. In this paper, we proposed a new algorithm using a modified Chi-squared test on clustered received signals as components to its performance function. Simulations show that even under low SNR environment, our proposed algorithm achieved higher recognition rate than other existing algorithms.