PathFinder: Advancing Path Loss Prediction for Single-to-Multi-Transmitter Scenario

TL;DR

PathFinder introduces an innovative deep learning architecture for radio path loss prediction that effectively models environmental features and multi-transmitter scenarios, improving generalization under distribution shifts.

Contribution

The paper presents PathFinder, a novel model with active environmental modeling, disentangled feature encoding, and a new benchmark for multi-transmitter prediction, addressing key limitations of prior methods.

Findings

PathFinder significantly outperforms existing methods in multi-transmitter scenarios.

The proposed approach demonstrates robust generalization under distribution shifts.

Experimental results validate the effectiveness of Mask-Guided Low-Rank Attention and Transmitter-Oriented Mixup.

Abstract

Radio path loss prediction (RPP) is critical for optimizing 5G networks and enabling IoT, smart city, and similar applications. However, current deep learning-based RPP methods lack proactive environmental modeling, struggle with realistic multi-transmitter scenarios, and generalize poorly under distribution shifts, particularly when training/testing environments differ in building density or transmitter configurations. This paper identifies three key issues: (1) passive environmental modeling that overlooks transmitters and key environmental features; (2) overemphasis on single-transmitter scenarios despite real-world multi-transmitter prevalence; (3) excessive focus on in-distribution performance while neglecting distribution shift challenges. To address these, we propose PathFinder, a novel architecture that actively models buildings and transmitters via disentangled feature encoding and integrates Mask-Guided Low-Rank Attention to independently focus on receiver and building regions. We also introduce a Transmitter-Oriented Mixup strategy for robust training and a new benchmark, single-to-multi-transmitter RPP (S2MT-RPP), tailored to evaluate extrapolation performance (multi-transmitter testing after single-transmitter training). Experimental results show PathFinder outperforms state-of-the-art methods significantly, especially in challenging multi-transmitter scenarios. Our code and project site are available at: https://emorzz1g.github.io/PathFinder/.