With the rapid development of science and technology, the Industrial Internet of Things (IIoT) has been improving and developing continuously since it was put forward, and the traditional industry has gradually moved towards networking and intellectualization. In the past, it was necessary to attach a radio frequency identification (RFID) label to each goods to complete the quantity monitoring, or to complete the counting by manpower. But the Radio Frequency Identification tags cannot be recycled, which will generate a lot of e-waste or increasing the cost of manpower. Therefore, in order to reduce the use of Radio Frequency Identification tags in practical applications, it is necessary to explore an innovative quantity monitoring system. We use the relationship between the quantity of goods and the digital signal to collect and analyze the data and information, and then to collect statistical data and real-time feedback information. The ultimate goal is to realize the intelligent management of goods in factory warehouse. In this paper, we propose a goods quantity monitoring system in a small warehouse. Firstly, we extract Radio Frequency (RF) signals in static and dynamic scene and preprocess them. Then, we extract the corresponding features according to different situations. Finally, we identify the quantity of goods according to K-Nearest Neighbors (KNN) classification algorithm. We have done a lot of experiments with Radio Frequency Identification equipment. The experimental results show that our system is robust and the average recognition accuracy reaches 95.53%.
Read full abstract