A method for judging customer actions in RFID clothing retail scenarios

In recent years, there have been notable developments in the field of the Internet of Things (IoT) with regard to Radio Frequency Identification (RFID) technology. Due to its low cost, passive operation, and long reading distance, RFID has demonstrated considerable potential for use in a variety of application scenarios. This paper puts forth a motion recognition method based on RFID technology, which captures subtle changes in RFID signals during different actions. By employing the Received Signal Strength Indicator (RSSI) and phase information, the system is able to accurately recognize human movements. The system employs two RFID readers, situated in disparate locations, and RFID tags integrated into clothing to discern actions. The system collects data on the variations in signal strength of the tags in response to different human actions, employs signal processing and feature extraction techniques, and uses machine learning algorithms to classify the actions. This paper introduces the use of Multi-Layer Perceptron (MLP) and K-Nearest Neighbors (KNN) algorithms to enhance the accuracy and robustness of the classification process. Experimental results show that the proposed algorithm achieves a classification accuracy of up to 87% and maintains satisfactory performance in complex environments.