Domain Adversarial Graph Convolutional Network Based on RSSI and Crowdsensing for Indoor Localization

In recent years, due to the wider WiFi coverage and the popularization of mobile communication devices, the technology of indoor positioning using WiFi fingerprints has been rapidly developed. Currently, most supervised methods need to collect a large amount of data to construct fingerprint datasets, which is labor-intensive and time-consuming. To solve the problem, we proposed a novel WiDAGCN model that can be trained with a few labeled site survey data and unlabeled crowdsensing WiFi fingerprints. To comprehensively represent the topology structure of the data, we constructed heterogeneous graphs according to the received signal strength indicators (RSSIs) between the waypoints and WiFi access points (APs). We focus on the graph convolutional network (GCN) method and the representation of graph-level features, which were rarely involved in previous WiFi indoor localization studies. Then, we try to minimize the difference between the source and target domains and make full use of the unlabeled data in the target domain using the domain adversarial training scheme. A public indoor localization dataset containing different buildings was used to evaluate the performance of the model. The experimental results show that our system can achieve a competitive localization accuracy in large buildings such as shopping malls.