Floor Recognition Based on SVM for WiFi Indoor Positioning

With the rapid development and popularization of WiFi technology, indoor positioning technology based on WiFi has become a research hot spot. At present, a variety of complex structures come into being along with the gradual improvement of people's living standards. The two-dimensional positioning of the room has been unable to meet people's needs. In this paper, a method of WiFi indoor positioning for floor recognition based on SVM (support vector machine) classification is proposed. Due to the obvious change during the WiFi signal through the wall or floor, floor identification is realized quickly and accurately by using SVM. We collect the RSS data of each floor with smart phone, and utilize the 10-fold cross-validation method to train classifier model and evaluate classification accuracy. The experimental results show that the high accuracy of floor discrimination, and the 99.09% floor recognition accuracy can be obtained.