Adaptive Indoor Localization System for Large-Scale Area

Generally, fingerprint-based indoor localization works inefficiently when deployed in a large-scale area. This is because it consumes massive resources and takes long processing time for searching the exact location large fingerprint database. Moreover, changing environment can degrade overall performance. To tackle these problems, we propose an adaptive system Our consists of three main parts. First, our area classification algorithm key to overcome problem caused by It identifies user's queries whether they are outdoor or located specific building. Specifically, filter out sent from out-of-scope areas. Then, information this part next part. Second, utilize first only significantly reduce space order localize location. Third, missing-BSSID detector detects missing Basic Service Set Identifiers (BSSIDs) incoming query updates sampling quickly adapt environment. We evaluated exhibition including 37 multi-floor buildings, covering 486,000 m 2 generating approximately 600,000 records users. In addition, created simulation evaluate critically-changing proposed achieves high accuracy. More importantly, compared previous work. Also, showed that applying as well other existing systems, performance be improved.