Uncertainty-Weighted Multi-Task CNN for Joint DoA and Rain-Rate Estimation Under Rain-Induced Array Distortions

TL;DR

This paper presents a multi-task deep CNN that jointly estimates direction-of-arrival and rain-rate from array data affected by rain distortions, improving accuracy over classical methods.

Contribution

It introduces an uncertainty-weighted multi-task CNN that automatically balances DoA and rain-rate estimation tasks under rain-induced array distortions.

Findings

Lower DoA RMSE than classical baselines

Accurate rain-rate classification at moderate-to-high SNRs

Effective joint estimation under rain-induced distortions

Abstract

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.