Signal modulation identification method of complex valued neural network based on structure optimization algorithm

Abstract

The invention discloses a signal modulation identification method of a complex valued neural network based on a structure optimization algorithm. The method comprises the following steps: initializing the structure and parameters of a forward complex valued neural network; adjusting parameters in the complex-valued neural network by using a complex-valued structure optimization algorithm, verifying the generalization performance of the forward complex-valued neural network, storing the number of current hidden layer neurons and all parameter values of the forward complex-valued neural network, and judging whether training is trapped in a local minimum point or not; if yes, adding a hidden layer neuron to the current model by using a complex value increment construction algorithm, calculating a weight of a newly added neuron, a hidden layer output matrix and a loss function value on the basis of current training, and continuing to train a new model; if not, keeping the model structure unchanged, and continuing training until an algorithm stopping condition is met. According to the method, the forward complex value neural network model with the optimal structure can be adaptively constructed, the method is used for identifying the communication signal modulation mode, and a satisfactory effect is achieved.

Description

technical field [0001] The invention relates to the fields of wireless communication technology and artificial intelligence, in particular to a communication signal modulation recognition method based on a structure optimization algorithm-based complex-valued neural network in a complex environment. Background technique [0002] The identification of the communication signal modulation mode is the prerequisite for obtaining the communication signal. Modulation recognition technology is a hot research topic in the field of signal processing in recent years. With the rapid development of communication technology, the modulation patterns of communication signals have become more complex and diverse, making it impossible for conventional identification methods and theories to effectively identify modulated signals, which also puts forward higher requirements for the identification research of modulation methods. [0003] Typical signal modulation recognition algorithms can be d...

AI Technical Summary

Problems solved by technology

However, the training speed, convergence speed and training results of the first-order complex-valued optimization algorithm are unsatisfactory, and there is currently no accurate mathematical model to solve the structural problems of the feed-forward complex-valued neural network

Images