A IoT Terminal Position Recognition Method Based on RSSI and Magnetic Field in A Low Trust Environment

With the rapid development of the IoT technology, a great amount of IoT terminals are put into use to improve intelligence of the power system. However, many IoT terminals are exposed in the low trust environment. They are easily to be moved or destroyed, which poses a great security risk. To determine whether an IoT terminal has been moved, we propose a position recognition method based on magnetic field and RSSI. Considering that the values of magnetic field vector change with the direction of the sensor, we use the magnetic vector calibration technology based on the rotation matrix to dynamically calibrate the vector sequence. To improve the effectiveness and efficiency in the scene which is small size or has low sample point densities, we design a neighborhood-based magnetic variation coefficient calculation strategy. In addition, we select effective reference points according to the standard deviation of signal strengths and propose an improved weighted KNN algorithm. By dynamically fusing the magnetic field vector and RSSI, and iteratively optimizing the weights, our method can effectively improve the recognition accuracy.