PROJECT TITLE :

SLAC: Calibration-Free Pedometer-Fingerprint Fusion for Indoor Localization - 2018

ABSTRACT:

To improve the accuracy of fingerprint-based mostly localization, one could fuse step counterwith fingerprints. However, the walking step model might vary among people. Such user heterogeneity may result in measurement error in walking distance. Previous works usually require a step counter tediously calibrated offline or through specific user input. Besides, as device heterogeneity could introduce numerous signal readings, these studies usually would like to calibrate the fingerprint RSSI model. Several of them haven't addressed how to jointly calibrate the higher than heterogeneities and find the user. We propose SLAC, a unique system which simultaneously localizes the user and calibrates the sensors. SLAC works transparently, and is calibration-free with heterogeneous devices and users. Its novel formulation is embedded with sensor calibration, where location estimations, fingerprint signals, and walking motion are jointly optimized with resultant consistent and proper model parameters. To cut back the localization search scope, SLAC 1st maps the target to a rough region (say, floor) via stacked denoising Autoencoders and then executes the fine-grained localization. Intensive experimental trials at our campus and also the international airport further make sure that SLAC accommodates device and user heterogeneity, and outperforms alternative state-of-the-art fingerprint-based mostly and fusion algorithms by lower localization errors (typically by additional than 30 p.c).