Research on Indoor Localization Algorithm Based on WIFI Signal Fingerprinting and INS

• Published in: 2018 International Conference on Intelligent Transportation, Big Data & Smart City (ICITBS) pp. 206 - 209
• Main Authors: Wang, An Yi, Wang, Lu
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.01.2018
• Subjects: accuracy; Big Data; Conferences; fingerprint; inertial navigation; neural network; Smart cities; Transportation; WIFI signal
• DOI: 10.1109/ICITBS.2018.00060

Using WIFI fingerprint positioning is Vulnerable to indoor environment and various noise signals , positioning error in indoor environment to the wireless signal strength is in the controllable range, although the starting point can be obtained, but its accuracy is easily affected by complex indoor environment. Inertial positioning is not affected by the indoor environment, but there is a cumulative error. At present, the widespread application of intelligent mobile phone, the activities of various sensors embedded can indirectly describe pedestrians, with intelligent mobile phone to achieve indoor positioning can not only reduce the cost of positioning, which can meet the daily demand for indoor positioning. Indoor positioning method based on inertial navigation real-time location of pedestrian path and wireless signal fingerprint with the online positioning by using the neural network matching algorithm outperforms the KNN algorithm in the positioning accuracy, so it can avoid the cumulative error of inertial navigation and positioning, and also reduce the influence of environment interference factor on the positioning accuracy, finally verified by experiments two, positioning algorithm combined with neural network positioning result is obviously better than the traditional positioning accuracy.