On-Device Learning of Indoor Location for WiFi Fingerprint Approach.

摘要: Indoor positioning is a recent technology that has gained interest in industry and academia thanks to the promising results of locating objects, people or robots accurately indoor environments. One utilized technologies based on algorithms process Received Signal Strength Indicator (RSSI) order infer location information without previous knowledge distribution Access Points (APs) area interest. This paper presents design implementation an mobile application, which allows users capture build their own RSSI maps by off-line training set selected classifiers using models generated obtain current target device. In early experimental stage, 59 were evaluated, data from proposed scenarios. Then, tested only top-five integrated with positioning, accuracy obtained for test The application achieves high classification rates, above 89%, at least 10 different locations environments, where each minimum separation 0.5 m.