The use of convolutional neural networks in radio tomographic imaging

This work aims to solve the problem of tracking people's movement in closed spaces. The applied solution does not require the monitored persons to have any devices with them. The method presented is to use radio tomographic imaging based on the fact that the human body is mostly water. This paper aims to show how heterogeneous and convolutional neural networks can be used to improve a radio tomographic imaging system that can accurately locate people indoors. In addition to the original algorithmic solutions, the advantages of the system include the use of properly designed and integrated devices -radio probes -whose task is to emit Wi-Fi waves and measure the strength of the received signal. Thanks to the two-step approach, the sensitivity, resolution and accuracy of imaging have increased. In addition, our solution performs well in radio tomography and other types of tomography because it is easy to understand and can be used in many ways.