OpenCSI: An Open-Source Dataset for Indoor Localization Using CSI-Based\n Fingerprinting

Abstract

Many applications require accurate indoor localization. Fingerprint-based\nlocalization methods propose a solution to this problem, but rely on a radio\nmap that is effort-intensive to acquire. We automate the radio map acquisition\nphase using a software-defined radio (SDR) and a wheeled robot. Furthermore, we\nopen-source a radio map acquired with our automated tool for a 3GPP Long-Term\nEvolution (LTE) wireless link. To the best of our knowledge, this is the first\npublicly available radio map containing channel state information (CSI).\nFinally, we describe first localization experiments on this radio map using a\nconvolutional neural network to regress for location coordinates.\n