Abstract: During the study of the technologies of energy saving, how to ensure the changing trend of traffic accurately is a prerequisite of many energy-saving technology. Contraposing the current methods for traffic prediction being a bit idealistic and the low accuracy prediction, we propose a traffic prediction method based on the seasonal autoregressive integrated moving average(S-ARIMA) model in view of the traffic character in the network and implement it. According to the characteristics of the traffic character in the network, We analyze the mathematical process of the S-ARIMA mode detailedly. It is tested by a lot of traffic data in the wireless communication networks and the results indicate that for prediction values and confidence intervals S-ARIMA model performs better than other models. Therefore, this traffic prediction for wireless communication networks using S-ARIMA model is reasonable and efficient.

Key words: seasonal autoregressive integrated moving average, time series, wireless communication networks, traffic prediction