Application of an Improved Neural Network Algorithm in Classification of Communication Noise Signals

Abstract

The genetic algorithm is used to improve the neural network and obtain the optimal weights and thresholds. Using the generated optimal weights and thresholds to run the neural network reduces the error norm over neural network which use random weights and thresholds. The classification of three kinds of communications noise signals was simulated by this algorithm, and the accuracy of this algorithm was proved. Experiments show that the algorithm will have an excellent application prospect in the noise signals classification in the communications field.