Experience: Practical Indoor Localization for Malls

Best Community Paper of Mobicom 2022

Yuming Hu, Feng Qian | Zhimeng Yin | Zhenhua Li | Zhe Ji, Qiang Xu, Yeqiang Han | Wei Jiang

University of Minnesota | City University of Hong Kong | Tsinghua University | XYZ10 Technology | SGCC

Abstract

We report our experiences of developing, deploying, and evaluating MLoc, a smartphone-based indoor localization system for malls. MLoc uses Bluetooth Low Energy RSSI and geomagnetic field strength as fingerprints. We develop efficient approaches for large-scale, outsourced training data collection. We also design robust online algorithms for localizing and tracking users’ positions in complex malls. Since 2018, MLoc has been deployed in 7 cities in China, and used by more than 1 million customers. We conduct extensive evaluations at 35 malls in 7 cities, covering 152K m^2 mall areas with a total walking distance of 215 km (1,100 km training data). MLoc yields a median location tracking error of 2.4m. We further characterize the behaviors of MLoc’s customers (472K users visiting 12 malls), and demonstrate that MLoc is a promising marketing platform through a promotion event.

System

MLoc consists of two phases: offline training, where (fingerprint, location) pairs are collected to build a localization model, and online inference, where a user’ smartphone collects fingerprints, uploads them to the edge, and obtains the location in real time. MLoc has been deployed since 9/2018 with improvements being made over the past three years. We have conducted principled, large-scale evaluations by hiring trained testers.

Beacon deployment

Fingerprint collecting

Corner cases of localization (narrow corridor, and dead end)

Dataset

We release the data used in the paper, most of which is collected in the past two years. MLoc adopts an outsourcing approach (i.e., hiring paid human workers) for collecting BLE/GMF fingerprints and the ground truth location data. The details of the data format can be found here, and an example of the trace file is shown here. We test MLoc based on the collected data, and the results are shown in the following figures.