Log Prediction of Wireless Telecommunication Systems Based on a Sequence-To-Sequence Model
Weiliang Ji
Department of Automation, University of Science and Technology of China, Hefei 230027, China.
Renai Chen
Department of Automation, University of Science and Technology of China, Hefei 230027, China.
Feng Li
Department of Automation, University of Science and Technology of China, Hefei 230027, China.
Qiang Ling *
Department of Automation, University of Science and Technology of China, Hefei 230027, China.
*Author to whom correspondence should be addressed.
Abstract
Nowadays people are becoming increasingly dependent on wireless networks. Taking precautions and acting in advance to avoid problems of wireless networks have already shown great importance. In analyzing problems of wireless telecommunication systems, current methods mainly rely on structured data, like alarm data. Compared with structured data, log data are more abundant and recently implemented to detect problems of wireless telecommunication systems. In order to predict future problems, it becomes essential to predict future log data based on current log data. In the paper, we propose a novel method to predict log data of the wireless telecommunication system based on a sequence-to-sequence model. We use current logs as input, and generate future log with our model. In addition we discuss the effects of different parameters of our model through some experimental results.
Keywords: Log prediction, sequence-to-sequence, machine learning, wireless telecommunication systems, machine log, sliding window