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

2026-02-04

AI Summary

This work proposes a Unified Joint Model (UJM) to address heterogeneous, non-Gaussian interference of unknown types in single-channel RF signals. UJM uniquely integrates interference detection and separation within a single deep neural network, departing from conventional Detect-Then-Separate (DTS) strategies that rely on multiple specialized models. Built upon a complex-valued UNet architecture and incorporating Gaussian mixture modeling with maximum a posteriori detection theory, UJM enables end-to-end joint optimization directly on baseband signals. Theoretical analysis establishes the asymptotic optimality of DTS as a performance benchmark. Experimental results demonstrate that UJM matches the performance of ideal DTS across diverse signal-to-interference-plus-noise ratios, interference types, and modulation orders, while maintaining robustness under mismatched training/test interference proportions, thereby exhibiting strong scalability and practical utility.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: State Estimation
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

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.

Problem

• RF signal separation
• interference uncertainty
• single-channel processing
• non-Gaussian interference
• signal recovery

Innovation

• Unified Joint Model
• Detect-Then-Separate
• RF Signal Separation
• Deep Learning
• Gaussian Mixture Model