Stratified sensor network self-calibration from TDOA measurements

This paper presents a study of sensor network calibration from time-difference-of-arrival (TDOA) measurements. Such calibration arises in several applications such as calibration of (acoustic or ultrasound) microphone arrays, bluetooth arrays, and radio antenna networks. We propose a non-iterative algorithm that applies a three-step stratification process, (i) using a set of rank constraints to determine the unknown time offsets, (ii) applying factorization techniques to determine transmitters and receivers up to unknown affine transformation and (iii) determining the affine stratification using the remaining constraints. This results in novel algorithms for direct recovery of both transmitter and receiver positions using TDOA measurements, down to 6 receivers and 8 transmitters. Experiments are shown both for simulated and real data with promising results.