Subjects: Information Science and Systems Science >> Basic Disciplines of Information Science and Systems Science submitted time 2023-02-14 Cooperative journals: 《桂林电子科技大学学报》
Abstract: Aiming at the problem that the sorting clustering positioning algorithm has low matching accuracy, and there are
abnormal fingerprint points in the fingerprint points used for position calculation, a Wi-Fi positioning algorithm with matching
optimization and distance assistance is proposed. According to the user's front and back position, distance and step
length, a matching deviation detection model is designed to determine the user's abnormal position and matching deviation;
the adjacent elements in the sorted received signal strength vector are compared with the set threshold to determine the
change position of the sorting feature vector of the point to be located, achieve the purpose of correction by exchange, and
obtain the corrected and merged class matching result; according to the distance between the user's position determined in the
time period m before the positioning and the fingerprint point in the matching class, the abnormal fingerprint points used for
position calculation are eliminated, so as to achieve more accurate indoor positioning. The simulation results show that the
class matching accuracy and the average positioning accuracy of the proposed algorithm are improved respectively by 17%
and 22%.