CS Paper Radar

Learning or Modeling? Virtual XL-RIS Observation Reconstruction for Near-Field Sensing

Tuo Wu, Jie Tang, Jiacheng Wang, Kangda Zhi, Maged Elkashlan, Matthew C. Valenti, Mérouane Debbah, Kenneth Tong

eess.SP ·submitted 2026-10-13 01:36 UTC · v1

Abstract

A passive extremely large-scale reconfigurable intelligent surface (XL-RIS) offers a position-sensitive aperture for near-field sensing, but its element-wise signals are not directly observed. Recovering this virtual RIS observation (VOR) from fewer base-station measurements requires prior information. This paper investigates when calibrated propagation geometry or paired training data provides the more effective reconstruction prior. For a common sensing operator and target signal, we develop off-grid near-field sparse parametric reconstruction (OG-NF-SPR) and compare it with compact and DenseNet-based deep reconstructors. OG-NF-SPR uses reflected-path projection for position initialization, followed by Bayesian estimation of all path gains and continuous position refinement. A causal extension incorporates temporal correlation without trajectory labels. Controlled simulations use near-field UAV tracking as an illustration and separately examine noise, position-distribution shift, calibration error, and propagation mismatch. Under matched propagation at 20 dB, OG-NF-SPR achieves approximately 67% lower reconstruction NMSE than the two nominal deep networks. At low SNR, statistical averaging is more effective than selecting a single geometric explanation. Under sufficiently strong repeatable mismatch, a same-size network trained on the mismatch family outperforms the nominal physical estimator, whereas nominal training does not provide this benefit. Temporal regularization also reduces position RMSE from 0.653 m to 0.225 m on a shared trajectory. The results relate reconstruction performance to measurement quality, calibration accuracy, and training coverage, providing guidance on when geometry can replace supervised labels and when additional data are useful.

Comment
13 pages, 6 figures

Closest papers

Nearest neighbours by meaning across the whole index.

Sparse Reconstruction by Direct Measurement-to-Field Generation in Function Space

cs.LG ·2026-10-26 ·Linzheng Wang, Jason Chen, Nicolas Tricard, Zituo Chen, Xingsen Guo +1 more
similarity 0.63
Abstract

Understanding physical systems often requires recovering their full spatial state from measurements available only at sparse locations. This reconstruction becomes challenging when physical boundaries are unknown, high-resolution training fields are unavailable, or important quantities cannot be measured directly. Sparse observations may also be consistent with multiple plausible states, leading to significant reconstruction uncertainty. Here we introduce Direct Measurement-to-Field Generation (\DMF{}), a functional rectified flow framework that learns how sparse measurements drive physical-field generation end to end. Instead of relying on grid-based rasterization or runtime guidance, \DMF{} jointly learns a sensor--global--sensor representation with the generative transport, dynamically fusing local grounding with global context to evaluate physical fields at arbitrary query coordinates. Across benchmarks involving unobserved boundaries, missing high-resolution training fields, and unmeasured physical quantities, \DMF{} outperforms alternative baselines by significantly reducing reconstruction errors and preserving the fidelity of physical properties. Furthermore, its stochastic ensemble generation reliably quantifies uncertainty while improving continuous ranked probability scores. These results extend sparse reconstruction from fixed-setting point estimation toward measurement-conditioned generation of physical fields under incomplete and changing information.

0.38

Wave-Domain Semantic ISAC: Holographic Surfaces as Analog Semantic Encoders for Integrated TN-NTN

cs.LO ·2026-11-10 ·Shadi Danesh, Pooria Salami, Hesaam Zaravashan, Abdulkadir Köse, Ayhan Akbas +2 more
published similarity 0.60
Abstract · 8 pages, 3 figures, accepted for publication in IEEE Communications Standards Magazine

Semantic communication, near-field operation, and integrated sensing and communication (ISAC) are often considered separately in terrestrial and non-terrestrial networks (TN--NTN). We show how a large electromagnetic aperture can connect them. \textcolor{black}{Nonlocal metasurfaces and holographic apertures act on the spatial spectrum of an incident field and can execute linear projections, which form an early stage of semantic feature extraction, directly in the wave domain before digitisation, thereby reducing the corresponding digital processing burden.} Near-field focusing also provides position-dependent sensing features, allowing the aperture to serve as an analog semantic-ISAC front end on resource-constrained aerial and satellite platforms. We describe the underlying physics, the co-design of constrained analog projections with lightweight digital decoders, and a case study in which a data-adaptive projection outperforms a fully digital baseline in the simulated energy-limited regime while transmitting four times fewer coefficients and using 11.4 times fewer decoder MACs. Finally, we identify relevant 3GPP, ITU-R, and ETSI activities and the remaining standardisation gaps.

0.56

TRACE: Retrospective Streaming Generation of Physical Fields under Sparse Structured Sensing

cs.LG stat.ML ·2026-08-26 ·Xinyu Zhang, Lihao Chen, Panqi Chen, Lei Cheng, Ting Zhang +2 more
similarity 0.59
Abstract · 8 pages, 4 figures, 1 table (main text); 11 figures, 15 tables in the 20-page appendix. Under review at AAAI 2027

Reconstructing continuous physical fields from sparse measurements is central to scientific monitoring, inverse modeling, and digital-twin construction. Generative reconstruction has recently emerged as a promising paradigm for this task by learning data-driven physical priors that complete plausible full fields from limited observations. However, existing methods largely assume fixed, batch conditioning, whereas real sensing systems often produce structured streams: probes scan local regions, instruments observe moving fields of view, and communication constraints may leave entire frames missing. We propose TRACE, a retrospective streaming generative reconstruction framework for physical fields under structured sensing. TRACE performs approximate Bayesian inference in a learned continuous-coordinate latent space, converting sparse off-grid measurements into generative latent evidence, fusing it with a state-space temporal prior through Kalman-style filtering, and refining under-observed past frames via retrospective smoothing. Experiments on active matter, ocean sound-speed fields, and supernova simulations show that TRACE matches or surpasses frame-wise generative reconstructors, offline spatiotemporal methods, and streaming data-assimilation baselines in reconstruction quality under temporally sparse and spatially localized sensing protocols.

0.20

Learning Field Reconstruction from Incomplete Data by Globally Correcting Local Estimates

cs.AI ·2026-10-04 ·Renhao Zhong, Zihan Zhou, Chiyuan Ma, Tianshu Yu
hot direction similarity 0.59
Abstract

Reconstructing physical fields from training samples that are always incomplete requires learning spatial structure from fragmented observations. Existing context--query work establishes how held-out observations provide valid training targets, but this does not make the complete-field distribution identifiable when every training field is incomplete. With finite data, weak evidence of sharp transitions and localized variations can further favor averaged predictions that attenuate local detail. A structural prior is therefore needed to favor plausible completions; local spatial relationships offer one grounded in the observations. We propose a locally constructed, globally revisable estimator that explicitly learns local field estimates and subsequently corrects them using full-domain observations. A shared coordinate-conditioned predictor learns from incomplete patches, allowing relatively well-observed neighborhoods to provide direct supervision of local structure. Its overlapping predictions are reconciled into an observation-conditioned consensus field. A full-domain estimator retains the original observations and learns a residual correction around this frozen field estimate, allowing locally constructed structure to be revised by broader evidence. The local estimate serves as both an explicit input, accompanied by its discrepancies with the observations, and a prediction starting point that the global model can revise. On three real-world ocean datasets with authentic observation gaps, our estimator achieves the lowest MSE and highest PSNR on withheld source-supported values, reducing MSE by 28.9%--34.5% against the strongest external baseline.

0.33

Low-Complexity Neural Pursuit for Line Spectral Estimation

eess.SP ·2026-09-19 ·Chenchen Liu, Wenjun Jiang, Xiaojun Yuan
top venuetheory similarity 0.59
Abstract

In this paper, we propose a high-precision yet low-complexity framework, termed Neural Pursuit for Line Spectral Estimation (NeuPLSE), for off-grid line spectral estimation from a single snapshot. The proposed framework systematically integrates model-driven signal structures with data-driven adaptability, enabling continuous-domain spectral representation together with adaptive parameter learning. Within this framework, model order detection and spectral parameter estimation are jointly achieved through residual analysis, with coordinated use of global and generalized component-wise residual information to ensure robust and accurate estimation performance. Moreover, NeuPLSE supports a unified extension from one-dimensional (1D) to multi-dimensional (MD) scenarios while maintaining low computational complexity. Extensive simulations in representative integrated sensing and communication (ISAC) scenarios demonstrate that NeuPLSE achieves accurate model order detection and high-precision spectral estimation. In sparse-path settings, it approaches the Cramér-Rao lower bound, while in dense-path settings, it attains competitive reconstruction performance with significantly lower computational complexity than state-of-the-art methods.

0.36

Inferring physical fields in coupled systems with unknown parameters from incomplete observations using physics-constrained attentive neural operators

cs.LG math.NA ·2026-10-05 ·Shilun Wei, Xiaoqiang Sun, Wei Li, Kejun Tang
similarity 0.59
Abstract

Given incomplete measurements of a single physical field in a coupled system with unknown parameters, can we infer its full physical state and identify the underlying parameters? This problem is challenging because multiple coupled fields must be reconstructed simultaneously from limited observations of only one, while the system parameters are unknown. In this work, we propose a machine learning framework for full-field reconstruction and parameter identification of unknown physical systems from sparse observations of a single physical field. Specifically, the cross-attention encoder propagates sparse sensor observations onto a regular grid to construct a sensor-conditioned latent representation, while a Fourier neural operator (FNO) decoder captures global spatial dependencies to reconstruct all coupled physical fields. The network parameters and unknown physical parameters are jointly optimized by minimizing observation losses, governing equation residuals, and boundary/initial condition constraints. The proposed approach is validated on two- and three-dimensional lid-driven cavity flows, a two-dimensional cylinder wake, and a two-dimensional non-ideal magnetohydrodynamics problem, demonstrating the recovery performance of unobserved fields and physical parameters from incomplete observations.

0.27

Near-Field Channel Estimation via Deep Unfolding for Extremely Large-Scale Antenna Arrays

eess.SP ·2026-11-09 ·Nguyen The Khang, Do Hai Son, Tran Trong Duy, Trung Thanh Le, Viet-Dung Nguyen +4 more
similarity 0.58
Abstract · 6 pages, 7 figures, submitted to the 2026 International Conference on Advanced Technologies for Communications (ATC 2026), Best Paper Award

The Extremely Large-scale Antenna Array (ELAA) has been envisioned as a promising candidate to meet the ambitious demands of 6G. However, this evolution also brings about two primary limitations: high computational complexity caused by large-scale arrays and inherent operation in the near-field region. To address these issues, this work exploits the sparsity of near-field channels in the polar domain and formulates channel estimation as a compressed sensing (CS) problem, thereby enabling estimation with reduced pilot overhead. In particular, we propose a guideline to design a sparsifying codebook for the near-field channel, which is strictly based on a target coherence value. Building on this, we develop a deep unfolding architecture, derived from the iterative solver ISTA, with a hybrid loss, namely Hybrid LISTA. Numerical results demonstrate that our proposed codebook, with comparable complexity, still exhibits improved estimation performance and greater coverage in the near-field region. Moreover, our unfolded model with the hybrid loss consistently delivers better recovery performance over existing estimation schemes.

0.32

Vision-Aided Near-Field Channel Estimation for XL-MIMO

eess.SP ·2026-11-04 ·Sachira Karunasena, Angel Lozano, Tom Drummond, Rajitha Senanayake
similarity 0.58
Abstract

In wideband extremely large-scale MIMO (XLMIMO) systems, the transmitter (TX) is often in the near field of the base station array. The resulting spherical wavefront makes the line-of-sight channel depend on both the TX angle and distance. Existing polar-domain estimators sample both parameters across the entire coverage region, requiring many candidate positions and pilots. This paper proposes a vision aided framework that uses the object carrying the TX, as seen in a camera image, to restrict the candidate positions to a small grid on that object. An on-grid vision-aided coarse localization (V-CL) algorithm obtains a coarse estimate from the grid points that best fit the received pilots, and an off-grid vision-aided Nelder-Mead refinement (V-NMR) algorithm refines it over continuous angle and distance, from which the channel is reconstructed. Simulations show that the proposed algorithms achieve higher accuracy than both wideband and narrowband polar-domain estimators while requiring far fewer pilots and candidate positions.

0.33