CAJ | 학술논문

针对当前网络流量预测模型精度低的缺点，本文提出了一种新型的小波消噪和蚁群算法优化支持向量机的网络流量预测模型。首先采用小波阈值法对网络流量进行消噪处理；然后将网络流量输入到支持向量机中学习，并采用蚁群算法对支持向量机的参数进行优化，建立网络流量预测模型，最后采用实际网络流量数据进行仿真实验，结果表明，相对于其它网络流量预测模型，本文模型提高了网络流量的预测精度，具有更好的鲁棒性。
침대당전망락류량예측모형정도저적결점，본문제출료일충신형적소파소조화의군산법우화지지향량궤적망락류량예측모형。수선채용소파역치법대망락류량진행소조처리；연후장망락류량수입도지지향량궤중학습，병채용의군산법대지지향량궤적삼수진행우화，건립망락류량예측모형，최후채용실제망락류량수거진행방진실험，결과표명，상대우기타망락류량예측모형，본문모형제고료망락류량적예측정도，구유경호적로봉성。
In order to solve the low prediction accuracy of network traffic prediction model, a new model of net-work traffic based on wavelet de-noising and support vector machine optimized by ant colony optimization algorithm is proposed in this paper. Firstly, the network traffic is processed by wavelet de-noising method, secondly, the data are input to support vector machine to learn which colony algorithm is used to optimize parameters of support vector ma-chine to establish the prediction model of network traffic, finally, the performance is test by network traffic data. The simulation results achieved with the actual traffic flow data show that the accuracy and the adaptability in the predic-tion are improved significantly which verifies that the proposed method is effective for network traffic prediction.