With the popularity of smart mobile devices, users can rely on global positioning system (GPS) technology when they are outdoors to determine their geographic location and to use other navigation services. However, GPS cannot reliably obtain satellite signals indoors, GPS is not suitable for indoor positioning applications. Wi-Fi and Bluetooth are the mainstream technologies used for indoor positioning today. But these radio technologies sometimes encounter problems, such as human-shadowing effects, multiple-path delays, and radio-wave interference, which can cause serious errors in the accuracy of indoor positioning results. In addition, using wireless network signals require the setup of certain infrastructure equipment. On the other hand, there are no such problems when using positioning technology based on the Earth's magnetic field. This paper analyzes how the environment can influence magnetic field measurements from magnetometers in mobile devices and verifies that this system can be used to enable indoor positioning. This paper also proposes the use of a weighted magnetic field component and the k-nearest neighbors (k-NN) algorithm for enhanced precision in indoor positioning. Finally, the research results show that a positioning accuracy of 91.7% and an average positioning error distance of 0.76 m can be achieved with these methods. The results of experiments show that the system performance is significantly feasible.