Massive multiple-input multiple-output (MIMO) is one of the promising technologies that can offer large capacities in multi-user scenarios with a large-scale antenna system. However, the base stations (BSs) consume too much energy when all of the antennas are turned on. If the users’ traffic requirements can be predicted, we may turn on/off antennas as needed to save energy while at the same time, guaranteeing users’ satisfaction. In this paper, we propose a clustering-based wavelet-LSTM method to predict the users’ traffic requirement in the next interval. According to the prediction results, we determine the number of antennas that needs to be turned on in the next interval. Our method is tested against a real-world anonymous dataset from an operator in a city in China. In comparison with some algorithms in machine learning, numerical results show that our clustering based wavelet-LSTM method achieves higher prediction precision. Furthermore, changing the on/off states of antennas by our proposed prediction method, we could get about 15% gain in energy consumption compared with the energy efficient system where states of antennas are adjusted by the number of users within the BS coverage.
[1]
Nan Liu,et al.
Prophet model and Gaussian process regression based user traffic prediction in wireless networks
,
2020,
Science China Information Sciences.
[2]
Xinlei Chen,et al.
Characterizing and Predicting Individual Traffic Usage of Mobile Application in Cellular Network
,
2018,
UbiComp/ISWC Adjunct.
[3]
Chao Zhang,et al.
DeepMove: Predicting Human Mobility with Attentional Recurrent Networks
,
2018,
WWW.
[4]
Basim Nasih,et al.
Application of Wavelet Transform and Its Advantages Compared To Fourier Transform
,
2016
.
[5]
Erik G. Larsson,et al.
Massive MIMO for next generation wireless systems
,
2013,
IEEE Communications Magazine.
[6]
Yoshua Bengio,et al.
Learning long-term dependencies with gradient descent is difficult
,
1994,
IEEE Trans. Neural Networks.
[7]
Ingrid Daubechies,et al.
The wavelet transform, time-frequency localization and signal analysis
,
1990,
IEEE Trans. Inf. Theory.