Calibration-Free Approaches for Robust Wi-Fi Positioning against Device Diversity: A Performance Comparison

Received signal strength (RSS) in Wi-Fi networks is commonly employed in indoor positioning systems; however, device diversity is a fundamental problem in such systems. This problem becomes more important in recent years due to the tremendous growth of new Wi-Fi devices, which perform differently in respect to the RSS values and degrade localization performance significantly. Several studies have proposed methods to improve the robustness of positioning systems against device diversity. This paper is primarily concerned with the performance of calibration-free approaches, including signal strength difference (SSD), hyperbolic location fingerprinting (HLF), and DIFF. The performance comparison is based on two Wi-Fi positioning systems in a 3-D indoor building, including a zero-configuration and a fingerprinting-based system. The results show that these calibration-free techniques perform much better than the original RSS with heterogeneous devices. However, the improvement in robustness is gained at the expense of losing some discriminative information. When the testing and training data are both measured from the same device, the performance of HLF and SSD is clearly below that of RSS in both systems. Although DIFF performs the best, it has to suffer from dealing with a space of large dimensions.