Multipath Parameter Optimization for Improving Accuracy of Ray-Tracing-Based Received Power Prediction in Industrial Environments

Wireless propagation channel modeling plays a crucial role in the design and planning of wireless networks. Ray-tracing (RT) method is a widely used site-specific approach to calculate the propagation channel for a given geometry of an environment. Due to the high-frequency approximation used in RT, it suffers from errors in the calculated received power especially at low frequencies where objects may be on the same order or smaller than the wavelength. In this paper, we propose to reduce this error by attenuating the calculated received power depending on the size of the reflecting or diffracting object with respect to wavelength. We also propose two approaches to perform optimization of the algorithm parameters using measured training data. The proposed algorithms are trained and evaluated using measurement data taken in an actual factory environment at five different center frequencies in the sub-6 GHz and mmWave bands. Results showed that the proposed algorithm managed to significantly reduce the RMSE of received power from 11.6 dB to 5.9 dB for the 922 MHz band, and to below 3 dB for the 2.4 GHz and higher bands.