Practical Passive Indoor Localization with Intelligent Reflecting Surface
Ganlin Zhang,Dongheng Zhang,Hongyu Deng,Yun Wu,Fengquan Zhan,Yan Chen
DOI: https://doi.org/10.1109/tmc.2024.3414861
IF: 6.075
2024-01-01
IEEE Transactions on Mobile Computing
Abstract:Intelligent reflecting surface has gained significant attention for supporting integrated sensing and communication (ISAC) by manipulating wireless signals. However, existing IRS-based sensing systems commonly utilize directional horn antennas for signal transmission, which is incompatible with commercial WiFi devices and contradicts the concept of ISAC. In this paper, we propose an IRS-aided localization system with omnidirectional antennas to overcome these limitations. We reveal two critical challenges for implementing such a system. Firstly, precise distance between the IRS and transmitter is requisite for IRS codebook design, yet practical measurement introduces centimeter-scale biases, causing severe localization errors. We derive a linear relationship between measurement bias and localization error, which serves as the basis for a calibration method. Secondly, in practice, we can only obtain the relative phase change of the IRS under different bias voltages, but cannot measure the phase offset introduced by IRS at zero voltage. Consequently, an unknown fixed offset appears when calculating the channel response of the signal reflected by the IRS, which destroys subsequent IRS codebook design. We resolve this problem by eliminating the signals that are not related to the IRS. Extensive experiments validate the proposed calibration techniques and demonstrate that our system achieves accurate passive localization.