Reciprocity-Aware Convolutional Neural Networks for Map-Based Path Loss Prediction

2025-04-04

AI Summary

Path loss models trained on downlink measurement data suffer from poor generalizability to uplink and backhaul links due to fundamental asymmetries in propagation environments and hardware configurations. Method: This paper proposes a single-data-source-driven, unified modeling framework for multi-link path loss prediction. Leveraging channel reciprocity as a physical prior, we embed it explicitly into a CNN architecture via map-feature extraction, physics-informed data augmentation, and a reciprocity-constrained loss function. The model is trained exclusively on downlink measurements, augmented with only a small number of synthetically generated uplink samples. Contribution/Results: Our approach enables joint modeling of downlink, uplink, and backhaul path loss without requiring dedicated uplink or backhaul measurements. Experiments demonstrate an RMSE reduction of over 8 dB on uplink test sets, while maintaining high accuracy across all three link types. This significantly enhances spectrum planning and cross-link interference coordination capabilities in integrated access and backhaul (IAB) networks.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Path loss modeling is a widely used technique for estimating point-to-point losses along a communications link from transmitter (Tx) to receiver (Rx). Accurate path loss predictions can optimize use of the radio frequency spectrum and minimize unwanted interference. Modern path loss modeling often leverages data-driven approaches, using machine learning to train models on drive test measurement datasets. Drive tests primarily represent downlink scenarios, where the Tx is located on a building and the Rx is located on a moving vehicle. Consequently, trained models are frequently reserved for downlink coverage estimation, lacking representation of uplink scenarios. In this paper, we demonstrate that data augmentation can be used to train a path loss model that is generalized to uplink, downlink, and backhaul scenarios, training using only downlink drive test measurements. By adding a small number of synthetic samples representing uplink scenarios to the training set, root mean squared error is reduced by>8 dB on uplink examples in the test set.

Problem

• Generalize path loss model to uplink, downlink, backhaul scenarios
• Improve accuracy using downlink drive test measurements only
• Reduce uplink prediction error via synthetic data augmentation

Innovation

• Reciprocity-aware CNN for path loss prediction
• Data augmentation generalizes model to uplink scenarios
• Synthetic samples reduce uplink prediction error