IoT Data Acquisition Node For Deep Learning Time Series Prediction

With the rapid development of Artificial Intelligence (AI) technology, its applications in the big data analytics and practices have become more and more extensive. Time series prediction based on deep learning, such as LSTM, GRU, and BiLSTM, is an important direction for future big data analytics and practices. In order to obtain the data needed for deep learning time series prediction more accurately and efficiently, Internet of Thing (IoT) technology is needed. In this paper, we propose a flexible, reliable and low-cost IoT data acquisition node. The node can collect the data required for deep learning time series data prediction efficiently for long time. The test results show that the node can effectively meet the design requirements and has the promise to provide a corresponding reference for future related designs.