Design and Evaluation of a Frequency Shift Backscatter Tag for Context Recognition

Abstract

Context recognition has attracted attention for various daily life applications, such as healthcare, fitness tracking, and elderly care. Many existing approaches for context recognition use micro-electro-mechanical systems (MEMS) sensors such as accelerometer and gyroscope. However, they require maintenance for charging or replacing batteries. In this paper, we propose a context recognition method using a frequency shift backscatter tag with ultra-low power consumption. The backscatter tag consists of an antenna, an oscillator, and a motion switch. The tag uses backscatter which leverages surrounding radio frequency (RF) waves emitted from an exciter for ultra-low power wireless communication. The tag changes the state of the motion switch according to the movement of humans or the change of the situation of things. The frequency of the backscattered signal is shifted according to the oscillation frequency. Context recognition is achieved by observing the existence of this frequency shift and its change over time. To demonstrate the feasibility of context recognition using the backscatter tag, we implemented a prototype and evaluated its performance. Our results demonstrate that our system using a BLE exciter can detect the frequency shift within 3 m.