The Research On The Scale Of The Mobile Communications Market's Client Based On Artificial Neural Network

China has the world's largest mobile communications market in the mobile communication network, especially the application of the 2G-2.5G communication technology, along with the issuance of 3G licenses, China Telecom, China Mobile, China Unicom and other operators face tremendous business opportunities,3G market will be the new battleground for telecom industry. How to accurately predict the growth of the scale of the users, plays a vital role in the decision-making for the operator.This artificial neural network technology to predict the scale of the field of mobile users, using MATLAB software to design prediction model for the size of mobile users based on BP neural network. By using the method of one related analysis of 12 predictors, as input of the model.Papers on the Levenberg-Marquardt method, gradient descent method, the gradient drive capacity factor decreasing law six different BP algorithms to compare the training effect analysis, the model for paper-Levenberg-Marquardt training algorithm algorithm. Finally, by changing the hidden nodes and network training times, the network training and prediction results were compared to determine the optimum hidden nodes was 11, Number of training times was 7, and the scale of this program as a mobile user forecasting model.Through testing sample set of the prediction model to verify the simulation results with the mainstream and the Rayleigh law curve model for comparison. The results show that mobile users based on the scale of BP network prediction model has good accuracy.