WiGrus: A Wifi-Based Gesture Recognition System Using Software-Defined Radio

Tao Zhang,Tian Zhang,Jie Zhuang,Daolin Chen,Tingyu Song

doi:10.1109/access.2019.2940386

Abstract

With the proliferation of WiFi devices and infrastructures, the ubiquitous WiFi signals are used to transmit user data. Besides it is also capable of sensing and identifying human gestures. In this paper, we propose a WiFi-based gesture recognition system, namely WiGrus, which solves the problems of user privacy and energy consumption compared with the approaches using wearable sensors and depth cameras. WiGrus leverages the fine-grained Channel State Information (CSI) extracted from WiFi signals to recognize a set of hand gestures. First of all, we utilize timestamps attached to the extracted CSI values to split continuously received WiFi packets into gesture instances. Second, a Principal Component Analysis (PCA)-based method and the first order difference are employed to reduce the noise and mitigate multipath effects caused by the environment changes. Then, massive features are extracted from the processed CSI values to present the intrinsic characteristics of each gesture. Finally, a 2-stage-RF algorithm is proposed to classify the gestures. Our experiments are implemented with a wireless router and a Software Defined Radio (SDR) device, more specifically Universal Software Radio Peripheral (USRP), which are used as WiFi signal transmitter and receiver respectively. The experimental results demonstrate that WiGrus can achieve an average accuracy of 96% in Line-of-sight (LOS) scenario and 92% in Non-Line-of-Sight (NLOS) scenario in the office environment and is robust to the environment changes.

Highlights

Human motion detection utilizes specific devices and approaches to extract the characteristics of a person’s movement states
We present a method that directly extracts the Channel State Information (CSI) values located in the preamble of Orthogonal Frequency Division Multiplexing (OFDM) frames, based on a modified IEEE 802.11g OFDM receiver
A wireless router (TP-Link TL-WR886N) with 3 antennas is fixed as the transmitter, and the Universal Software Radio Peripheral (USRP) N210 with one SBX daughterboard as well as one antenna (VERT2450) is fixed as the receiver which is connected to a laptop installed with GNU Radio [35]

Summary

INTRODUCTION

Human motion detection utilizes specific devices and approaches to extract the characteristics of a person’s movement states. Many researchers focus on sensors to sense and detect human motions Those methods require people to be equipped with dedicated sensors, such as motion sensors, accelerometer sensors, and gyroscopes, to collect movement information [1]–[4]. Kinect [6], can detect human motions with exceedingly high accuracy This camera-based method only works well in the line-of-sight (LOS) scenario. With the ubiquity of WiFi devices and infrastructures, WiFi-based motion detection methods attract the interests of considerable researchers, e.g., [9]–[13], which solve the problems of privacy as well as requirements for specific environments or sensors. Experimental results manifest that our algorithm is superior to other classification algorithms with an average accuracy of 96% in LOS, and 92% in NLOS scenarios

RELATE WORK

COMMERCIAL HARDWARE-BASED GESTURE RECOGNITION

DATA PREPROCESSING Data preprocessing mainly contains two steps

Background

CONCLUSION