Long-Short Term Memory Networks For Resource Allocation Forecasting in Wifi Networks
暂无分享,去创建一个
Long Short Term Memory (LSTM) networks have been proven to be a remarkably useful model for time series forecasting. However, there are some elements such as seasonality, holiday effects and regressors which lead to difficulties in the forecasting process. In this paper, we propose a LSTM model with a data preprocessing method called Min-max scaling [7] and apply on predicting the number of views in a public wifi system, based on this number the manager can optimize resource management. The numerical simulation with real data are reported to demonstrate the efficient of our approach.
[1] W. Fuller,et al. LIKELIHOOD RATIO STATISTICS FOR AUTOREGRESSIVE TIME SERIES WITH A UNIT ROOT , 1981 .
[2] W. Fuller,et al. Distribution of the Estimators for Autoregressive Time Series with a Unit Root , 1979 .
[3] Yoshua Bengio,et al. Learning long-term dependencies with gradient descent is difficult , 1994, IEEE Trans. Neural Networks.
[4] Jürgen Schmidhuber,et al. Long Short-Term Memory , 1997, Neural Computation.