Indoor Localization Based on CFR Environment Awareness

Accurate amplitude information provides the possibility for indoor localization, it is easy to acquire and does not require complex corrections like phase. However, the amplitude information is sensitive to the environment, which reduces the localization accuracy. This paper proposes an indoor localization method based on environment awareness. First, using the variance of phase difference to judge whether there is anyone walking in the environment. Second, we create amplitude images as the input to Convolution Neural Network without the person walking to train localization model in the offline. Third, online localization using localization model.