Research On Interactive Behavior Recognition And Optimization Based On Wireless Sensing

With the recent surge of the Internet of things technology and the rapid development of human recognition technology,the research of human-computer interaction has attracted wide attention from the academic community.Unlike the traditional switch or console commands,Using arm gestures to control devices has more advantages such as natural and convenient.Wireless sensing based action recognition does not require users to wear extra devices,and is little affected by environmental factors like lighting or sound,which is of great significance to provide efficient services in smart home,industrial production and other special areas.As a mature wireless sensing device,Doppler radar detects the motion information in the environment with stable electromagnetic frequency,which provides a possibility for high precision and high reliability human arm gesture recognition.In this paper,by studying Doppler radar signal data,we design and implement a human arm gesture recognition system based on Doppler radar with the following three main contributions:(1)The paper designs a gesture type recognition model based on Doppler radar.Combining with the features of Doppler radar signals,the paper proposes a layout method to eliminate detection blind areas by two vertical placed radar.Based on this,the paper proposes a high accuracy gesture type identification approach with a two-stage classification method.(2)To achieve a fine-grained gesture sensing,the paper proposes a gesture detail analysis approach with a spectrum analysis method,including two practical calculation ways such as extremum ratio method and area ratio method,to obtain the detailed information such as the angle,amplitude or direction of arm movements.(3)To adapt the system to different environments,the paper proposes multi-optimization methods including synchronization optimization,multi-person environment optimization and processing efficiency optimization in the recognition model,to reduce deployment cost and noise interference,improve efficiency and enhance system stability.This paper collects radar signals about arm movements from multiple users in various scenarios,and verifies the correctness and reliability of the proposed method,which provides a new technical support for intelligent human-computer interaction services.