LAUSR

Exploring and developing the perception system is an important means to promote the development of intelligent transportation. In this paper, we propose RAPP, a radio tomography-based method enabling performance parameterization to achieve parameter adjustment of localization accuracy without dense deployment of passive UHF RFID devices. The reader antenna takes a rapid-moving system as the carrier and generates virtual antenna elements to inventory received signal strength (RSS) of the pre-deployed tags. To eliminate the impact of the lack of tag backward signals during the mobile process and supplement the link distribution for each segment of the monitored platform, the Cubic Spline Interpolation algorithm (CSIA) is proposed to homogenize the spacing of the virtual antenna elements by preprocessing the measurements. Then the optimized link information is adopted to perform radio tomographic imaging to recognize the locations of targets. Theoretical analysis on ill-condition is derived to analyze how the number of virtual interpolation elements affects the localization accuracy. Further derivation about the relationship between the Bayesian Cramér-Rao Bound (CRB) and the link density provides a solution to influence the localization performance through parameter adjustment. Extensive simulation and experiment show that RAPP outperforms on multiple targets localization and behaves robustly in a multipath environment.