Domain clustering based WiFi indoor positioning algorithm

This paper focuses on WiFi indoor positioning based on received signal strength, a common local positioning approach with a number of prominent advantages such as low cost and ease of deployment. Weighted k nearest neighbor (WKNN) approach and Naive Bayes Classifier (NBC) method are two classic position estimation strategies for location determination using WiFi fingerprinting. Both of them need to handle carefully the issue of access point (AP) selection and inappropriate selection of APs may degrade positioning performance considerably. To avoid the issue of AP selection and hence improve positioning accuracy, a new WiFi indoor position estimation strategy via domain clustering (DC) is proposed in this paper. Extensive experiments are carried out and performance comparison based on experimental results demonstrates that the proposed method has a better position estimation performance than the existing approaches.