Vortex EM Wave-Based Rotation Speed Monitoring on Commodity WiFi

Abstract

In industry, various machinery involves spinning components, and monitoring their rotation speed is important for predicting the operation state. There are many approaches for rotation speed estimation, each with its own limitations, e.g., traditional contact methods, requiring specialized equipment, non-contact methods, such as optical methods affected by illumination and occlusion, and acoustic methods, susceptible to environmental noise. Radio frequency signals like wireless fidelity (WiFi) can provide non-intrusive approach for rotation sensing. However, conventional WiFi sensing struggles to capture self-spinning objects when there is no radial motion towards or away from the WiFi transceivers. This paper proposes a rotation speed estimation method based on vortex electromagnetic (EM) waves. By using a q-shaped patch antenna on commodity WiFi, vortex EM waves with spiral phase front can be transmitted in WiFi communication channel. When these vortex EM waves interact with the surface of a self-spinning object, the echos exhibit a frequency shift compared to the original wave. This frequency shift is proportional to the object's rotation speed, and is known as the rotational Doppler effect. This paper first derives the mathematical relationship between WiFi channel state information (CSI) and rotational Doppler frequency shift (RDS), and we design a series of signal processing methods to extract RDS from WiFi CSI, achieving the rotation speed estimation of self-spinning targets. Experimental results show that this method is applicable to a rotation speed range of 0–7000 rpm, with an relative error within 5%.