Spatial Correlation in Shadow Fading for Wireless Communication

Signal propagation in wireless communication undergoes an effect called shadowing (or shadow fading) depending on the built and natural surroundings that impede propagation. Thus, the fixed surroundings contribute to a level of correlation in shadow fading over a given distance in space, which is referred to as spatial correlation in shadow fading. This paper propose a method to simulate homogeneous spatial-correlation in shadow fading in the 2D space with a roughly specified decorrelation distance, using a 2D convolutional process having a 2D linear time invariant system. Such methods are commonly used in image processing. Simulation results are presented to verify the gener ated correlated spatial process for the shadowing component, and further insights are drawn through 2D spatial autocorrelation values. The proposed method for a single transmitter can be repeated for multiple transmitters that have its own shadowing signal component, when simulating a network. The introduced methods are particularly useful for simulated data generation for the validation of machine learning and deep learning model that facilitate next generation wireless communications.