Skip to main content

Deep Learning Propagation Models over Irregular Terrain

01 January 2019

New Image

Accurate path gain models are critical for coverage prediction and RF planning in wireless communications. In many settings irregular terrain induces blockages and scattering with great impact on path gain, making its prediction a challenging task. Current solutions for these environments are either computationally expensive or slope-intercept fits that cannot capture local deviations due to terrain variation, leading to large prediction errors. We propose to use machine learning to learn path gain based on terrain altitudes as features. We implement different neural network architectures with dense and convolutional layers that could include effects difficult to describe with traditional models (e.g. back scatter). We test our framework on an extensive set of measured path gain data and consistently predict with 5 dB RMSE, an 8 dB improvement over traditional slope-intercept solutions.