Saved in:
Bibliographic Details
Main Authors: Yan, Chenyang, Yang, Ruonan, Sun, Shunqiao, Bengtsson, Mats
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.01961
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911416327340032
author Yan, Chenyang
Yang, Ruonan
Sun, Shunqiao
Bengtsson, Mats
author_facet Yan, Chenyang
Yang, Ruonan
Sun, Shunqiao
Bengtsson, Mats
contents We investigate joint direction-of-arrival (DoA) and rain-rate estimation for a uniform linear array operating under rain-induced multiplicative distortions. Building on a wavefront fluctuation model whose spatial correlation is governed by the rain-rate, we derive an angle-dependent covariance formulation and use it to synthesize training data. DoA estimation is cast as a multi-label classification problem on a discretized angular grid, while rain-rate estimation is formulated as a multi-class classification task. We then propose a multi-task deep CNN with a shared feature extractor and two task-specific heads, trained using an uncertainty-weighted objective to automatically balance the two losses. Numerical results in a two-source scenario show that the proposed network achieves lower DoA RMSE than classical baselines and provides accurate rain-rate classification at moderate-to-high SNRs.
format Preprint
id arxiv_https___arxiv_org_abs_2602_01961
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Uncertainty-Weighted Multi-Task CNN for Joint DoA and Rain-Rate Estimation Under Rain-Induced Array Distortions
Yan, Chenyang
Yang, Ruonan
Sun, Shunqiao
Bengtsson, Mats
Signal Processing
We investigate joint direction-of-arrival (DoA) and rain-rate estimation for a uniform linear array operating under rain-induced multiplicative distortions. Building on a wavefront fluctuation model whose spatial correlation is governed by the rain-rate, we derive an angle-dependent covariance formulation and use it to synthesize training data. DoA estimation is cast as a multi-label classification problem on a discretized angular grid, while rain-rate estimation is formulated as a multi-class classification task. We then propose a multi-task deep CNN with a shared feature extractor and two task-specific heads, trained using an uncertainty-weighted objective to automatically balance the two losses. Numerical results in a two-source scenario show that the proposed network achieves lower DoA RMSE than classical baselines and provides accurate rain-rate classification at moderate-to-high SNRs.
title Uncertainty-Weighted Multi-Task CNN for Joint DoA and Rain-Rate Estimation Under Rain-Induced Array Distortions
topic Signal Processing
url https://arxiv.org/abs/2602.01961