Extended LTE Based Fingerprinting Positioning for Emergency Applications by Utilizing Seq2seq Model with Beam-Search Inference

Outdoor location-based services (LBS) not only provide convenience in daily life, but also requires location information that operates in real time and has high accuracy and availability for stable rescue activities of rescuers and rescuers, especially in emergency rescue situations. Global Positioning System (GPS) or Global Navigation Satellite System (GNSS) provide an accurate and reliable localization performance under line of sights wireless environments. But there remains a problem that signals cannot be received indoors or in densely populated urban areas, or location estimation errors increase due to multi-path signals. In this work, in order to increase the number of available LTE base stations, which is a key factor in the accuracy of fingerprinting LBS, we collect LTE parameters of all three telecommunication companies providing services in Korea by mobile collection device. Because CID is an identifier that includes the meaning of PCI and EAFRCN, we select CID among LTE parameters to predict available base stations which constructs pseudo cell. In order to improve the accuracy of the previous work [11], the existing SGNS (Skip-Gram Negative Sampling) model was changed to the Seq2Seq model and a beam search method suitable for Seq2Seq was applied to the inference part of the DL (Deep Learning) model instead of the commonly used greedy search.