Prediction for Wireless Traffic with Long Range Burstiness

Abstract

Recently many studies have found that the self-similarity in nature brings about the long range burstiness of the traffic across the wide range of time scales. The updated experimental evidences reveal that the heavy tailness is the key cause for the self-similar of the network traffic. Future fourth-generation (4G) wireless networks will be characterized by a 3G core network architecture. The multimedia services will be provided to mobile users, the 3G core network will have to support aggregated traffics, exhibiting long range dependent (LRD) and self-similar characteristics, as it occurs in the Internet. We deal with a forecasting method melding three predictors including autoregressive (AR), moving average (MA) and fractional autoregressive integrated moving average (FARIMA) based on alpha-stable innovation for the wireless traffic with long range burstiness according to the minimum dispersion criteria with infinite variance. The final predicted values are obtained by combining the previous three individual predicted values. The predicted results for the actual traces show that the three individual predictors are precise and effective, the last compound predictors can enhance the final predicted accuracy.