Deep learning targets clutter in shared wireless sensing

Wireless signals used for communication can also support sensing. A preprint by Yalin E. Sagduyu and colleagues explores whether deep learning can help extract target information when those signals are obscured by clutter.

What the study found

The authors present a multistatic integrated sensing and communications framework, in which spatially distributed receivers contribute to sensing using shared wireless infrastructure and spectrum. They model clutter using randomly located scatterers with varying radar cross sections and Doppler spreads, representing differences in how the scatterers reflect signals and in their motion-related frequency shifts.

According to the authors, conventional signal-processing approaches showed limited robustness under strong clutter. Their residual and attention-enhanced neural networks, designed to enhance target responses in clutter-contaminated observations, substantially outperformed those approaches in target position and velocity estimation across the clutter conditions studied.

Why it matters

Dense or dynamic clutter can interfere with estimates of range and Doppler, introducing errors when determining where a target is and how fast it is moving. Better clutter suppression could therefore improve the usefulness of multistatic sensing alongside communications. The study frames distributed receivers as a way to increase sensing coverage, while investigating a limitation that can undermine estimation.

Limits and unknowns

The available source is the abstract of an arXiv preprint, not a full account of the evaluation. It gives no numerical performance results, names no individual baseline methods and does not specify the training data, evaluation setup or hardware. It also does not establish how the methods would perform in real-world deployments or under conditions outside those studied.

Read the research

Authors: Yalin E. Sagduyu, Tugba Erpek, Sastry Kompella

Published: 2026-10-29

Source basis: Based on the abstract only

Original abstract

Integrated sensing and communications (ISAC) systems using 5G and beyond signals of opportunity enable joint wireless sensing and communication using shared infrastructure and spectrum resources. While multistatic ISAC can increase sensing coverage with spatially distributed receivers, dense and dynamic clutter can significantly degrade bistatic range and Doppler estimation, leading to errors in target localization and velocity estimation. This paper presents a multistatic ISAC framework with clutter suppression for high-fidelity estimation of target position and velocity. Distributed clutter is modeled with randomly located scatterers with varying Doppler spreads and radar cross sections to capture dense and dynamic clutter effects. A broad range of conventional signal processing baselines are evaluated for clutter suppression, showing limited robustness under strong clutter conditions. To address this limitation, deep learning-based clutter suppression methods are introduced to improve robustness against dense and dynamic clutter by employing residual and attention-enhanced deep neural network architectures that directly enhance target responses from clutter-contaminated observations. These methods substantially outperform conventional approaches in target position and velocity estimation across diverse clutter conditions in multistatic ISAC.

This article is an AI-generated adaptation, not a substitute for the original research. Preprints may not have been peer reviewed; AI can make mistakes.