Evaluation of Communication Countermeasure Combat Capability Based on SOM-BP Cloud Neural Networks
Shi Jun-ta · Fire Control and Command Control · 2013
In order to enhance reliability and accuracy of neural network training samples,to improve self-learning and self-adapting ability of network,the evaluation result is made quickly and accurately. In this paper,according to the communication countermeasure combat capability index,on the basis of cloud-neural network,a evaluation method is presented base on SOM-BP cloud neural networks. This method uses SOM clustering for sample data,training the improved BP cloud-neural network by choosing the characteristic sample. The result shows that this method not only has faster convergence,fewer iterations,but also has the higher accuracy rate than BP cloud neural network.