Base Station Network Traffic Prediction Approach Based on LMA-DeepAR

TL;DR

This paper introduces LMA-DeepAR, a novel deep learning model that improves long-term base station network traffic prediction by incorporating local moving average features to handle non-stationary and bursty traffic patterns.

Contribution

The paper proposes a new traffic prediction model combining LMA feature extraction with DeepAR, enhancing accuracy and stability for non-stationary network traffic.

Findings

LMA-DeepAR outperforms existing methods in long-term prediction accuracy.

The approach effectively reduces interference from non-stationary traffic.

Experimental results demonstrate improved stability in multi-cell network traffic prediction.

Abstract

Accurate network traffic prediction of base station cell is very vital for the expansion and reduction of wireless devices in base station cell. The burst and uncertainty of base station cell network traffic makes the network traffic nonlinear and non-stationary, which brings challenges to the long-term prediction of network traffic. In this paper, the traffic model LMA-DeepAR for base station network is established based on DeepAR. Acordding to the distribution characteristics of network traffic, this paper proposes an artificial feature sequence calculation method based on local moving average (LMA). The feature sequence is input into DeepAR as covariant, which makes the statistical characteristics of network traffic near a period of time in the past be considered when updating parameters, and the interference of non-stationary network traffic on model training will be reduced. Experimental results show that the proposed prediction approach (LMA-DeepAR) outperforms other methods in the overall long-term prediction performance and stability of multi cell network traffic.