Model-based interference cartography and visualization

Abstract

In this work, we present a tool to construct and visualize the spatio-temporal variations of power. A dataset of real-world power measurements is collected over a geographical area of interest. Relevant parameters of the environment such as the path loss exponent and the decorrelation time of the lognormal shadow fading are extracted from the dataset. Also, the average powers measured at a finite set of known locations are interpolated to obtain the average power distribution over the area. Using the parameters of the lognormal shadow fading, synthetic data with the same temporal behavior of the dataset is generated, and multiplied with the average power distribution. The resulting spatio-temporal power map is displayed on the screen through a graphical user interface developed in-house. The proposed approaches for interpolation and parameter extraction are validated using test datasets generated using the well-accepted modified Gudmundson model for the spatio-temporal correlation of lognormal shadow fading. We also undertake a comparative study of three different interpolation techniques: linear interpolation, inverse distance weighing and ordinary kriging. Further, we compare a model-based approach with a model-free approach for interpolation, and find that model-based ordinary kriging provides the best mean absolute percentage error performance.