Application of Improved Ant Colony Algorithm in Indoor Location

摘要

Due to geomagnetic properties being disturbed by the environment, the accuracy of indoor location based on geomagnetic field is lower. The paper improves the ant colony algorithm by adding elite strategy during pheromones update in indoor location, and then the positioning accuracies of a series of ant colony algorithms and KNN method are compared by experiments, the results of which show: (1) the improved algorithm is superior to the other methods based on the different combinations of parameters, especially for the easily confused and disturbed locations. (2) KNN method is suitable for the following occasions: the attributes of different locations are independent, and the attributes of the same location at different times are consistent. (3) The geomagnetic properties of different locations at different times are subject to diverse interference so that the positioning accuracies are also affected; the positioning accuracies of working hours are lower than that of non-working hours, and the positioning accuracies of the clean locations in a quiet environment are higher.