Learning methods for RSSI-based geolocation: A comparative study

In this paper, we investigate machine learning approaches addressing the problem of geolocation. First, we review some classical learning methods to build a radio map. These methods are split in two categories, which we refer to as likelihood-based methods and fingerprinting methods. Then, we provide a novel geolocation approach in each of these two categories. The first proposed technique relies on a semi-parametric Nadaraya–Watson (NW) estimator of the likelihood, followed by a maximum a posteriori (MAP) estimator of the object's position. The second technique consists in learning a proper metric on the dataset, constructed by means of a Gradient boosting regressor: a k-nearest neighbor algorithm is then used to estimate the position. The proposed methods are compared on two data sets originated from Sigfox network, and an indoor dataset performed in a three-story building. Experiments show the interest of the proposed methods, both in terms of location estimation performance, and ability to build radio maps.