Learning to Separate RF Signals Under Uncertainty: Detect-Then-Separate vs. Unified Joint Models

TL;DR

This paper compares detect-then-separate strategies with a unified joint model for RF signal separation, demonstrating that a single neural network can effectively handle diverse interference types and outperform traditional methods in complex, uncertain environments.

Contribution

The paper introduces a unified deep learning model for RF signal separation that matches the performance of traditional detect-then-separate methods, improving scalability and practicality.

Findings

UJM matches oracle DTS performance across various conditions.

UJM handles type-uncertainty and mismatched training/testing effectively.

DTS is a principled benchmark but less scalable.

Abstract

The increasingly crowded radio frequency (RF) spectrum forces communication signals to coexist, creating heterogeneous interferers whose structure often departs from Gaussian models. Recovering the interference-contaminated signal of interest in such settings is a central challenge, especially in single-channel RF processing. Existing data-driven methods often assume that the interference type is known, yielding ensembles of specialized models that scale poorly with the number of interferers. We show that detect-then-separate (DTS) strategies admit an analytical justification: within a Gaussian mixture framework, a plug-in maximum a posteriori detector followed by type-conditioned optimal estimation achieves asymptotic minimum mean-square error optimality under a mild temporal-diversity condition. This makes DTS a principled benchmark, but its reliance on multiple type-specific models limits scalability. Motivated by this, we propose a unified joint model (UJM), in which a single deep neural architecture learns to jointly detect and separate when applied directly to the received signal. Using tailored UNet architectures for baseband (complex-valued) RF signals, we compare DTS and UJM on synthetic and recorded interference types, showing that a capacity-matched UJM can match oracle-aided DTS performance across diverse signal-to-interference-and-noise ratios, interference types, and constellation orders, including mismatched training and testing type-uncertainty proportions. These findings highlight UJM as a scalable and practical alternative to DTS, while opening new directions for unified separation under broader regimes.