摘要
Location-Based Services have become an indispensable part of our daily life, the sparsity of location finding makes it possible to estimate specific position by Compressive Sensing(CS). Using public Frequency Modulation(FM) broadcasting and Digital Television Terrestrial Multimedia Broadcasting(DTMB) signals, this paper presents an indoor positioning scheme, which is consisted of an offline stage and an online stage. In the offline stage, the Received Signal Strength(RSS) at the Reference Points(RPs) is measured, including the average and variance of public FM broadcasting and DTMB signals. In the online stage, the K-Weighted Nearest Neighbor algorithm is used to fulfill coarse positioning, which is able to narrow the selection scope of locations and choose partial RPs for accurate positioning. Then, the accurate positioning is implemented through CS. Experiment shows that the average positioning error of the proposed scheme is 1.63 m. What’s more, a CS-based method has been proposed in this paper to reduce the labor cost when collecting data. Experiment shows the average positioning error is 2.04 m, with the advantage of a 34% labor cost reduction. Experiment results indicate that the proposed system is a practical indoor positioning scheme.
Location-Based Services have become an indispensable part of our daily life, the sparsity of location finding makes it possible to estimate specific position by Compressive Sensing(CS). Using public Frequency Modulation(FM) broadcasting and Digital Television Terrestrial Multimedia Broadcasting(DTMB) signals, this paper presents an indoor positioning scheme, which is consisted of an offline stage and an online stage. In the offline stage, the Received Signal Strength(RSS) at the Reference Points(RPs) is measured, including the average and variance of public FM broadcasting and DTMB signals. In the online stage, the K-Weighted Nearest Neighbor algorithm is used to fulfill coarse positioning, which is able to narrow the selection scope of locations and choose partial RPs for accurate positioning. Then, the accurate positioning is implemented through CS. Experiment shows that the average positioning error of the proposed scheme is 1.63 m. What’s more, a CS-based method has been proposed in this paper to reduce the labor cost when collecting data. Experiment shows the average positioning error is 2.04 m, with the advantage of a 34% labor cost reduction. Experiment results indicate that the proposed system is a practical indoor positioning scheme.
基金
supported in part by the National Natural Science Foundation of China under grant No.61571244 and No.61871239
in part by Tianjin Research Program of Application Foundation and Advanced Technology under grant No.16YFZCSF00540 and No.18YFZCGX00480
