A Self-Calibrated Localization System using Chirp Spread Spectrum in a Wireless Sensor Network

To achieve accurate localization information, complex algorithms that have high computational complexity are usually implemented. In addition, many of these algorithms have been developed to overcome several limitations, e. g., obstruction interference in multi-path and non-line-of-sight (NLOS) environments. However, localization systems those have complex design experience latency when operating multiple mobile nodes occupying various channels and try to compensate for inaccurate distance values. To operate multiple mobile nodes concurrently, we propose a localization system with both low complexity and high accuracy and that is based on a chirp spread spectrum (CSS) radio. The proposed localization system is composed of accurate ranging values that are analyzed by simple linear regression that utilizes a Big-O(n(2)) of only a few data points and an algorithm with a self-calibration feature. The performance of the proposed localization system is verified by means of actual experiments. The results show a mean error of about 1 m and multiple mobile node operation in a 100x35 m(2) environment under NLOS condition.