CITYMPC: A Large-Scale Physics-Informed Benchmark and Tool for Generative Complete Multipath Wireless Channel Modeling

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Arun, Ashwin Natraj, Nickel, David R., Zhang, Yaguang, Xing, Yunchou, Chen, Jie, Ghosh, Amitava, Brinton, Christopher, Love, David J., Krogmeier, James V.
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909045973057536
author Arun, Ashwin Natraj
Nickel, David R.
Zhang, Yaguang
Xing, Yunchou
Chen, Jie
Ghosh, Amitava
Brinton, Christopher
Love, David J.
Krogmeier, James V.
author_facet Arun, Ashwin Natraj
Nickel, David R.
Zhang, Yaguang
Xing, Yunchou
Chen, Jie
Ghosh, Amitava
Brinton, Christopher
Love, David J.
Krogmeier, James V.
contents Multipath wireless channels are fully characterized by multipath components (MPCs), including complex channel gain, propagation delay, angle of departure (AoD) and angle of arrival (AoA) in azimuth and elevation. Generating these parameters with the fidelity of ray tracing (RT) remains an open problem. Existing methods either incur the computational cost of RT or require explicit 3D scene geometry at inference. We present CITYMPC, a conditional variational autoencoder (cVAE) that predicts the complete per-path MPC parameter set from point-of-view imagery and terrain height maps alone, achieving environment-aware channel generation without access to any three-dimensional scene geometry at inference. Trained and evaluated across five urban environments spanning 427,397 links, CITYMPC matches RT ground truth to within 1.29 dB received power mean absolute error (MAE) and 7.25 ns $τ_0$ MAE. CITYMPC is a generative channel modeling framework and reproducible benchmark, released together with a large-scale multi-city ray-traced dataset to accelerate future scene-conditioned channel modeling research. We further analyze cross-city distribution shift to characterize the per-city diversity of the benchmark.
format Preprint
id arxiv_https___arxiv_org_abs_2605_15471
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle CITYMPC: A Large-Scale Physics-Informed Benchmark and Tool for Generative Complete Multipath Wireless Channel Modeling
Arun, Ashwin Natraj
Nickel, David R.
Zhang, Yaguang
Xing, Yunchou
Chen, Jie
Ghosh, Amitava
Brinton, Christopher
Love, David J.
Krogmeier, James V.
Signal Processing
Multipath wireless channels are fully characterized by multipath components (MPCs), including complex channel gain, propagation delay, angle of departure (AoD) and angle of arrival (AoA) in azimuth and elevation. Generating these parameters with the fidelity of ray tracing (RT) remains an open problem. Existing methods either incur the computational cost of RT or require explicit 3D scene geometry at inference. We present CITYMPC, a conditional variational autoencoder (cVAE) that predicts the complete per-path MPC parameter set from point-of-view imagery and terrain height maps alone, achieving environment-aware channel generation without access to any three-dimensional scene geometry at inference. Trained and evaluated across five urban environments spanning 427,397 links, CITYMPC matches RT ground truth to within 1.29 dB received power mean absolute error (MAE) and 7.25 ns $τ_0$ MAE. CITYMPC is a generative channel modeling framework and reproducible benchmark, released together with a large-scale multi-city ray-traced dataset to accelerate future scene-conditioned channel modeling research. We further analyze cross-city distribution shift to characterize the per-city diversity of the benchmark.
title CITYMPC: A Large-Scale Physics-Informed Benchmark and Tool for Generative Complete Multipath Wireless Channel Modeling
topic Signal Processing
url https://arxiv.org/abs/2605.15471