A Digital Modulation Recognition Method Based on Uniform Linear Array

Abstract

Classical digital modulation recognition methods will suffer from performance deterioration when SNR decreases, a method based on Uniform Linear Array (ULA) was proposed to solve above problem in this paper. The SNR of received signal will be increased by spatial filtering using ULA, and then the performance of modulation recognition at low SNR will be improved. The proposed modulation recognition method employes ULA as signal receiver, uses high-order cumulants as classification features, and finally Back-propagation Neural Network (BPNN) is used for digital modulation classification. Simulation results showed that the proposed method can recognize typical digital modulations including 2ASK, 4ASK, 4FSK and 4PSK efficiently with satisfactory recognition rate at low SNR. And meanwhile influence factors including number of array elements, structure of neural network and data scale of training and testing were studied and discussed.