cs.DCApr 11, 2026

RF-LEGO: Modularized Signal Processing-Deep Learning Co-Design for RF Sensing via Deep Unrolling

Authors: Luca Jiang-Tao Yu, Chenshu Wu

Organizations: The University of Hong Kong

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Wireless sensing, traditionally relying on signal processing (SP) techniques, has recently shifted toward data-driven deep learning (DL) to achieve performance breakthroughs. However, existing deep wireless sensing models are typically end-to-end and task-specific, lacking reusability and interpretability. We propose RF-LEGO, a modular co-design framework that transforms interpretable SP algorithms into trainable, physics-grounded DL modules through deep unrolling. By replacing hand-tuned parameters with learnable ones while preserving core processing structures and mathematical operators, RF-LEGO ensures modularity, cascadability, and structure-aligned interpretability. Specifically, we introduce three deep-unrolled modules for critical RF sensing tasks: frequency transform, spatial angle estimation, and signal detection. Extensive experiments using real-world data for Wi-Fi, millimeter-wave, UWB, and 6G sensing demonstrate that RF-LEGO significantly outperforms existing SP and DL baselines, both standalone and when integrated into multiple downstream tasks. RF-LEGO pioneers a novel SP-DL co-design paradigm for wireless sensing via deep unrolling, shedding light on efficient and interpretable deep wireless sensing solutions. Our code is available at https://github.com/aiot-lab/RF-LEGO.

Figures & tables

Figure 1. Core principles of RF-LEGO : Modularity, Cascadability, and Interpretability. RF-LEGO bridges the gap between classical SP and DL for RF sensing via deep unrolling.
Figure 2. Deep Unrolling. S : Signal Processing Block; O : Unrolled Trainable Operator; I : Unrolled Trainable Iterative Block.
Figure 3. RF-LEGO FT. (a) Bluestein’s Algorithm implementation of Fourier transformation via convolution, where b\boldsymbol{b} represents the chirp signal and z∗\boldsymbol{z^{*}} denotes its conjugate. (b) RF-LEGO FT replaces the fixed convolution with a learnable convolutional layer.
Figure 4. RF-LEGO Beamformer. (a) The classical LASSO solver using ADMM. (b) The unrolled architecture with learnable parameters. Here, x(t)\boldsymbol{x^{(t)}} represents the estimated sparse angular spectrum, z(t)\boldsymbol{z^{(t)}} is the auxiliary variable, v(t)\boldsymbol{v^{(t)}} is the dual variable, and g(t)\boldsymbol{g^{(t)}} denotes the learned gate that dynamically balances the update between historical and current estimates.
Figure 5. RF-LEGO Detector. (a) Classical CFAR uses a fixed sliding window. (b) RF-LEGO unrolls the logic into a state space model, where r\boldsymbol{r} is the input signal vector, z\boldsymbol{z} represents the learned state vector, and s^\boldsymbol{\hat{s}} is the detection vector derived from the state.
Figure 6. Experimental scenarios. (a-f) Range experiments; (g-h) Doppler experiments; (g,i) Angle experiments.
Figure 7. The results of RF-LEGO Range FT of mmWave and RF-LEGO Doppler FT of mmWave, UWB, and Wi-Fi.
Figure 8
Figure 11. Microbenchmarks. (a) Performance on public datasets. (b) Impact of multiple targets. (c) Impact of data fine-tuning.
Table 1. Ablation study.
Table 2. Inference latency on different edge platforms. The results are formatted as RF-LEGO / SP [ms].
Figure 12. Effect of interface sensitivity under cascading.
Figure 13. Module behavior analysis. (a) RF-LEGO FT learns a non-uniform convolution kernel beyond Bluestein’s Algorithm FT. (b) RF-LEGO Beamformer learns adaptive gate schedules across unrolled iterations. (c) RF-LEGO Detector learns a structured state matrix A\boldsymbol{A} , unlike memoryless CFAR.
Figure 14
Figure 15
Figure 16
More in this topic

Deep Unfolding

Deep unfolding turns iterative algorithms into trainable networks, combining model-based structure with learned components for efficient inference and optimization.

See all 37 papers

Explore similar work

CardsList

  1. Deep Learning-Enhanced Real-Time Wi-Fi Sensing Through Single Transceiver Pair

    Oct 21, 2025Yuxuan Liu, Chiya Zhang, Yifeng Yuan +3Wi-Fi SensingWireless Localization

    School of Electronic and Information Engineering, Harbin Institute of Technology, Shenzhen 518055, China. · School of Electronic and Information Engineering, Harbin Institute of Technology, Shenzhen 518055, China, and also with the Peng Cheng Laboratory (PCL), Shenzhen 518055, China. · Guangdong Key Laboratory of Intelligent Information Processing, Shenzhen University, Shenzhen 518060, China. +1

  2. RF-GPT: Teaching AI to See the Wireless World

    Feb 16, 2026Hang Zou, Yu Tian, Bohao Wang +4Wireless CommunicationsLanguage Modeling

    Research Institute for Digital Future, Khalifa University, 127788 Abu Dhabi, UAE · College of Information Science and Electronic Engineering, Zhejiang University, 310027, Hangzhou, China · Université de Lorraine, CNRS, CRAN, F-54000 Nancy, France

  3. An Uncertainty-Driven Hybrid Deep Learning Approach for Broad-Coverage RF Modulation Recognition

    Aug 1, 2026Nurettin Safak, Durdu Can Yerdeyatar, Muhammet Sefa Demirel +3Automatic Modulation Classification

    Department of Electrical and Electronics Engineering, Ankara Yildirim Beyazit University, Ankara, Turkey