A Novel FMCW Radar-Based Scheme for Indoor Localization and Trajectory
Tracking
Xiangyu Shen
College of Physics and Information Engineering
Fuzhou University
Fuzhou, Fujian, China
e-mail: xyshen1998@163.com
Haifeng Zheng
College of Physics and Information Engineering
Fuzhou University
Fuzhou, Fujian, China
e-mail: zhenghf@fzu.edu.cn
Xinxin Feng
College of Physics and Information Engineering
Fuzhou University
Fuzhou, Fujian, China
e-mail: fxx1116@fzu.edu.cn
Abstract—In recent years, with the rapid development of the
Internet of Things (IoT) related technologies, the location-
based service (LBS) has attracted increasing attention.
Frequency modulated continuous wave (FMCW) is a
promising technique for indoor localization due to its low
deployment costs and strong anti-interference ability. In this
paper, we develop a novel FMCW radar-based positioning
scheme for indoor localization and target trajectory detection.
In particular, we propose an OSCA-CFAR algorithm with
star-shaped reference sliding window to improve the accuracy
of indoor positioning. Furthermore, the extended Kalman filter
is introduced into the scheme to solve the nonlinear problem of
motion trajectory tracking. Finally, we carry out extensive
experiments with a real testing platform. Experimental results
demonstrate that the proposed scheme is able to achieve better
performance compared with the existing schemes.
Keywords- Indoor localization, FMCW radar, trajectory
tracking, CFAR, extended kalman filter
I. INTRODUCTION
With the rapid development of IoT related technologies,
the Location-Based Service (LBS) industry has witnessed
rapid growth. However, due to the building shielding and
other factors, the highly mature satellite positioning
technology in the outdoor environment cannot provide high-
precision positioning information in the indoor environment
[1]. At present, indoor positioning technology is widely used
in indoor navigation, fire site rescue, smart home, and many
other fields [2].
In order to improve the positioning accuracy of the
indoor environment, many indoor positioning technologies
have appeared in recent years, such as Wi-Fi positioning,
Radio Frequency Identification (RFID), and Bluetooth and
ultrawideband (UWB) radio positioning. Besides the above
positioning techniques, frequency modulated continuous
wave (FMCW) has become a promising positioning
technique due to its many advantages in anti-jamming and
positioning accuracy [3].
In recent years, many efforts have been made to improve
FMCW radar-based positioning performance. For example,
Katabi et al. [4] [5] developed WiTrack, an indoor
positioning system that uses human radio reflection for 3D
tracking. By receiving signals reflected from a moving
human body and conducting 3D modeling to determine the
human body’s position, the accuracy can reach 10-21cm. Jia
et al. [6] proposed an indoor stationary human body
detection algorithm based on FMCW radar, which uses life
signal detection and ellipse cross-location to locate human.
Wang et al. [7] proposed a hybrid radar scheme that
combines interferometric radar and FMCW radar for indoor
positioning. Xiong et al. [8] also proposed a linear FMCW
radar-based system with an improved constant-false-alarm
rate (CFAR) algorithm for indoor positioning and moving
target trajectory tracking.
However, the current works still face some challenges.
On one hand, due to the numerous objects, walls, etc, there
are various noises caused by multi-path interference in the
indoor environment, which are mixed in the echo signal
received by the receiving antennas reducing the positioning
accuracy of FMCW radar. On the other hand, the current
works on FMCW radar-based indoor positioning have not
efficiently solved the problem of the moving targets’
trajectory tracking. Since a target’s motion is usually
nonlinear, if it does not consider this characteristic in the
trajectory estimation algorithm, it is easy to bring cumulative
errors and cannot achieve high-precision performance for
trajectory tracking.
In this paper, we develop an FMCW radar-based
positioning scheme for indoor localization and target
trajectory detection to tackle the above issues. The main
contributions of this paper are summarized as follows:
x To eliminate the influence of background noise on
positioning accuracy, we propose the OSCA-CFAR
algorithm with star-shaped reference sliding window.
298
2020 IEEE 6th International Conference on Computer and Communications
978-1-7281-8635-1/20/$31.00 ©2020 IEEE
2020 IEEE 6th International Conference on Computer and Communications (ICCC) | 978-1-7281-8635-1/20/$31.00 ©2020 IEEE | DOI: 10.1109/ICCC51575.2020.9345047
Authorized licensed use limited to: Nanjing Univ of Post & Telecommunications. Downloaded on February 19,2021 at 09:44:55 UTC from IEEE Xplore. Restrictions apply.