Joint modulation recognition method based on RBM networks and BP neural network

Abstract

The invention discloses a joint modulation recognition method based on RBM networks and a BP neural network. The joint modulation recognition method comprises the steps that 1, a modulation signal ispreprocessed; 2, feature parameters are extracted; 3, a training sample and a test sample of each modulation mode are randomly generated; 4, an input layer and a first hidden layer of the BP neural network are taken as one RBM network to be trained; 5, a collection of parameters of the RBM network is initialized; 6, the RBM network is trained to obtain the collection of the parameters and output features of the hidden layer; 7, the first hidden layer and a next layer of the BP neural network are taken as a visible layer and a hidden layer of a second RBM network to be trained, the output of the first RBM network is taken as the input of the second RBM network, and the step 5, the step 6 and the step 7 are repeatedly conducted until collections of parameters of all RBM networks are acquired; 8, the BP neural network is retrained until an optimal solution state is achieved; and 9, test data is normalized, the trained BP neural network is input, and the modulation mode recognition rate iscalculated. The joint modulation recognition method has the advantages that the input dimension is reduced, and the modulation recognition rate is increased.

Description

technical field [0001] The invention relates to a modulation recognition method, in particular to a joint modulation recognition method based on an RBM network and a BP neural network, and belongs to the technical field of communication. Background technique [0002] With the development of communication technology, the application of signal modulation identification covers almost the entire commercial and military communication fields, and plays an important role in signal authentication, interference identification, electronic countermeasures, etc., and has a very wide range of application value and prospects. [0003] Modulation identification refers to the identification of the modulation mode before the demodulation of the received signal. Generally, there are three methods for modulation identification: decision tree classifier, cluster analysis, and neural network classifier. [0004] Nandi and Azzouz proposed a decision tree algorithm for classification based on fea...

AI Technical Summary

Problems solved by technology

Cluster analysis is a method of multivariate statistical classification, which performs blind classification based on the pattern similarity in unlabeled samples. However, the cluster analysis method is easily affected by noise, and the different extracted feature parameters have a greater impact on the recognition performance. Big

As the most common methods, backpropagation (back propagation, BP) and radial basis (radial basis function, RBF) neural networks with self-learning and generalization capabilities are very suitable for classification problems with potential nonlinear mappings between input signals and outputs. However, the neural network classification method is prone to fall into the local optimal solution problem.

In addition, the neural network classification method has a slow convergence rate when it is close to the optimal solution, and its generalization ability is poor, and the recognition rate is low when the signal-to-noise ratio (SNR) is low.

Images