Compressive Sensing Wireless Channel Modeling With Digital Map

Channel model plays a very important role in the wireless communication systems for verification of the key technologies in the mobile communications, such as modulation, coding and antenna design, etc. Usually, the channel model can be achieved by experimental method, which needs a large amount of measured data. In order to save the cost, some electromagnetic methods have been proposed to simulate the channel model, such as ray tracing and finite-difference time domain (FDTD). These methods can provide a good approximation of the real channel. However, in the traditional way, the uniform sampling is adopted, and the amount of simulated data is quite large. It costs a great deal of calculation and memory resources, which results in a time-exhausted computation. In this letter, compressive sensing (CS) is employed to decrease the complexity in channel modeling with simulation by digital map. The sparsity of the receiving signal is revealed and highlighted, which leads to the fact that the signal can be measured randomly less than the traditional Nyquist sampling rate. The simulation confirms that the complexity can be reduced at least 90% with root mean square error (RMSE) less than 10 $^{-2}$ .