基于遗传神经网络的P2P流量识别系统

P2P traffic recognition system based on genetic neural network

考虑到传统BP神经网络在进行P2P流量识别时,具有系统识别速度慢、精度低,神经网络自身容易陷入局部最小值等问题,使用遗传算法对BP神经网络进行优化。遗传算法具有较强的自适应性和鲁棒性,因此使用遗传算法对BP神经网络的权值和阈值进行优化处理,能够有效提高神经网络的性能。建立基于遗传神经网络的识别系统,采集处理大量样本数据,对识别系统进行训练和测试。研究结果表明,基于遗传神经网络的P2P流量识别系统具有识别精度高、识别速度快等优点,相比传统BP神经网络,其识别性能有明显提高。

Since the traditional BP neural network has slow recognition speed and low accuracy while proceeding P2 P traffic recognition,and the neural network itself is easy to fall into local minimum value,the genetic algorithm is used to optimize BP neural network. Genetic algorithm has better adaptability and robustness,so it is used to optimize the weight and threshold of BP neural network,which can improve the performance of the neural network effectively. To study the performance of the established recognition system based on genetic neural network,the recognition system was trained and tested by collecting and processing a large number of sample data. The research results show that P2 P traffic identification system based on genetic neural network has high recognition accuracy and rapid recognition speed. Compared with the traditional BP neural network,the recognition performance of P2 P traffic identification system has been improved obviously.