Inertial measurement unit–aided dual-frequency radio frequency identification localization in line-of-sight and non-line-of-sight hybrid environment. (March 2018)

• Record Type: Journal Article
• Authors: Wu, Jie; Zhu, Minghua; Xiao, Bo; He, Wei
• Is Part Of: International journal of distributed sensor networks. Volume 14:Number 3(2018)
• Publication Date: 2018-03
• Subjects: Dual-frequency radio frequency identification -- indoor localization -- non-line-of-sight -- received signal strength -- inertial measurement unit; Sensor networks -- Periodicals; Intelligent agents (Computer software) -- Periodicals; Multisensor data fusion -- Periodicals; 681.2
• DOI: 10.1177/1550147718762033
• Languages: English
• ISSNs: 1550-1329

Abstract

The mitigation of non-line-of-sight propagation conditions is one of main challenges in wireless signal–based indoor localization. When radio frequency identification localization technology is applied in applications, the received signal strength fluctuates frequently due to the shade and multipath effect of radio frequency signal, which could result in localization inaccuracy. In particular, when tag carriers are walking in line-of-sight and non-line-of-sight hybrid environment, great attenuation of received signal strength will happen, which would result in great positioning deviation. The article puts forward a dual-frequency radio frequency identification–based indoor localization approach in line-of-sight–non-line-of-sight hybrid environment with the help of inertial measurement unit. Dual-frequency radio frequency identification includes passive radio frequency identification and active radio frequency identification. Passive radio frequency identification is used to assist in determining the tag initial location with passive reader. Active radio frequency identification is used to locate the tag and send the sensor information to active radio frequency identification readers. The proposed method includes three improvements over previous received signal strength–based positioning methods: inertial measurement unit–aided received signal strength filtering, inertial measurement unit–aided line-of-sight/non-line-of-sight distinguishing, and inertial measurement unit–aided line-of-sight/non-line-of-sight environment switching. Also, Cramér–Rao low bound is calculated to prove theoretically that indoor positioning accuracy for the proposed method in line-of-sight and non-line-of-sight mixed environment is higher than position precision using only received signal strength information. Experiments are conducted to show that the proposed method can reduce the mean positioning error to around 3 m without site survey.