Path Loss Prediction Using Deep Learning

2024-11-25

arXiv.org

AI Summary

To address the low prediction accuracy and reliance on hand-crafted features in path loss modeling for wireless communication deployment, this paper proposes an end-to-end, path-level path loss prediction method. The method takes high-resolution Digital Surface Models (DSMs) as input and employs Convolutional Neural Networks (CNNs) to directly learn discriminative spatial features from obstacle distributions, eliminating dependence on manually engineered metrics such as equivalent terrain height. It establishes the first fully supervised regression framework that requires no pre-extracted obstacle statistics, thereby significantly enhancing cross-scenario generalizability. Evaluated across diverse urban and suburban real-world measurement sites, the method achieves over 40% reduction in mean absolute error compared to classical empirical models. This enables high-accuracy, scalable, and automated path loss prediction—supporting robust spectrum planning and base station placement.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Learning on the Edge & Model Compression
• Intelligent Robots: Motion and Path Planning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Radio deployments and spectrum planning benefit from path loss predictions. Obstructions along a communications link are often considered implicitly or through derived metrics such as representative clutter height or total obstruction depth. In this paper, we propose a path-specific path loss prediction method that uses convolutional neural networks to automatically perform feature extraction from high-resolution obstruction height maps. Our methods result in low prediction error in a variety of environments without requiring derived metrics.

Problem

• Wireless Signal Prediction
• Path Loss
• Spectrum Planning

Innovation

• Convolutional Neural Network
• Signal Attenuation Prediction
• High-definition Obstacle Height Maps