ZU-mean: fingerprinting based device localization methods for IoT in the presence of additive and multiplicative noise

This paper proposes Zero-Mean and Unity-Mean (ZU-Mean) features based device localization methods for internet of things (IoT). These features do not depend on the hardwares and/or specifications of the devices being used. Moreover, the zero-mean and unity-mean features mitigate the additive and multiplicative noise, respectively. Extensive real experiments are conducted in two different sites (residential and mall areas) using WiFi received signal strength (RSS) for five weeks. The performance of the proposed methods is better than the absolute RSS based method. We also highlight that the absolute RSS feature cannot be used in calibration-free method and hence, it is not suitable for diverse devices in IoT networks. Additionally, the proposed low-cost method is computationally efficient as compared to the existing methods in the literature.