WIFI Indoor Location Optimization Method Based on Position Fingerprint Algorithm

GPS based location technologies can achieve high accuracy on meter level in outdoor environments, however, it cannot obtain high level position accuracy in the indoor environment. Therefore, in this paper, we propose a novel WIFI indoor location optimization approach based on position fingerprint algorithm. Each fingerprint contains a set of intensity values of the detected WAPs, and then it is defined as a vector with fixed size. The proposed location fingerprinting algorithm is made up of online and offline phase, in which generating the location fingerprint database is a crucial issue. Particularly, at the end of training process, a probabilistic map is generated for the interest area, and a single probability density function for the access point and the reference point are defined as well. Finally, we choose three smartphones with different hardware configurations on Android OS to conduct an experiment. Experimental results show that the proposed algorithm can effectively optimize the WIFI indoor location.