Pith. sign in

REVIEW 5 major objections 6 minor 81 references

GLocFM: A Geometry-Aware Foundation Model for 3D Indoor Wireless Localization

T0 review · 5 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read GLocFM claims that a single-forward-pass model using a 3D scene point cloud to predict delay-angle spectra can localize a WiFi transmitter to 0.1429 m mean error in unseen rooms and 0.0637 m on real measurements, roughly halving the error…

desk verdict A competent, useful empirical paper whose headline gains are plausible but whose geometric-prior story is overclaimed; worth reviewing seriously once code/data and variance reporting appear. read the letter →

arxiv 2608.09285 v1 pith:IN6YAQCH submitted 2026-08-10 eess.SP cs.AI

Chenghong Bian , Chaozheng Wen , Hongze Chen , Jun Zhang This is my paper · ORCID
classification eess.SPcs.AI
keywords indoorlocalizationgeometry-awaremodeldelay-anglespectrumraytracingWiFisensing3DpointcloudToFrobustness
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper sets out to show that a wireless localizer that explicitly reasons about a room's 3D geometry can outperform learning-only approaches and generalize to rooms it has never seen. The authors build GLocFM, which turns localization into a maximum-likelihood problem over a grid of candidate transmitter positions, scoring each candidate by comparing the observed WiFi delay-angle spectrum against a spectrum predicted from direct and one-bounce paths through the scene point cloud. Because the exact likelihood is intractable, a learned scorer refines a cosine-similarity prior, and a small off-grid network produces a continuous position. Trained on 221 procedurally generated indoor scenes with ray-traced WiFi signals, the model reports mean 3D errors of 0.1429 m with four receivers and 0.1746 m with three receivers on unseen synthetic scenes, 49.5% and 56.0% below the strongest baseline, and 0.0637 m on the real-measurement NeRF2 dataset, 48.8% below the same baseline. If these results hold, they matter because indoor localization is a common problem where GPS fails, and a single forward pass with scene geometry would be both practical and generalizable.

What carries the argument

The load-bearing object is the predicted delay-angle-of-arrival spectrum for each candidate transmitter position. For a candidate and a receiver, the model sums the array responses of the LoS path and one-bounce reflection paths, where each path contributes a Gaussian-smoothed delay peak, an array steering vector, and a learned attenuation that includes distance loss, opacity-based soft visibility, and a reflection MLP. Comparing this predicted spectrum with the measured spectrum through cosine similarity produces a per-candidate prior, and the prior is fed into a learned scoring function whose output is converted into a softmax grid score plus an off-grid correction. This two-stage design lets the geometry do the coarse ranking while the learned part resolves ambiguities and achieves sub-grid resolution.

What would settle it

Use the Sionna RT full simulation with interaction depth three to compute, for each test scene, the fraction of received power carried by paths with two or more bounces, and check whether GLocFM's per-scene mean error rises with that fraction; a clear positive trend would confirm that the LoS-plus-one-bounce prior is the limiting assumption rather than the learned scorer.

Watch

Extended reading notes

Core claim

The paper's central claim is that matching measured radio observations against geometry-predicted spectra is enough to localize a transmitter in 3D with centimeter-to-decimeter accuracy in unseen rooms. Concretely, GLocFM partitions the free space of a scene into a grid, and for each grid center it synthesizes the expected delay-angle spectrum from the line-of-sight path and one-bounce reflection paths, using a point-cloud scene encoder, a learned opacity-based soft visibility for obstruction, and a small MLP for reflection attenuation. The measured channel is converted into the same delay-angle representation, and the cosine similarity between the two acts as a geometric prior. A transformer-based scorer fuses these similarities across receivers and learns to predict a probability per grid cell, while a bounded off-grid component refines the output to a continuous position. This pipeline is trained end-to-end on a synthetic dataset of 221 scenes, and is reported to cut mean error by roughly half relative to the strongest LocGPT baseline on both unseen synthetic scenes and the real-measurement NeRF2 dataset; a ToF-robust variant trained with random clock offsets keeps error stable when an unknown time offset is present.

Load-bearing premise

The approach assumes that almost all received signal power travels from transmitter to receiver either along a straight line or after reflecting off exactly one surface, so paths with two or more bounces can be ignored when predicting the radio spectrum.

Editorial extensions

If this is right

  • A single forward pass through a geometry-aware model can beat methods that run optimization at inference time, so real-time indoor localization with centimeter-level accuracy becomes plausible.
  • Using reflected paths as geometric constraints makes the model robust to losing one receiver: with three receivers the mean error rises only from 0.1429 m to 0.1746 m, while baselines degrade by roughly 0.1 m.
  • Because the ToF-robust variant keeps its mean error between 0.2044 and 0.2270 m for time offsets up to 4 ns without estimating the offset, the approach can work in unsynchronized deployments.
  • The model retains much of its accuracy when fine-tuned on only 25% of real training data, suggesting the synthetic pretraining transfers to real measurements.
  • The ablations show that the candidate grid and the learned scorer are essential, with a coarse grid raising mean error from 0.1429 m to 0.4494 m and removing the learned scorer raising it to 0.3914 m.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The reported gains likely come from a combination of the geometric prior and the learned scorer; an ablation that replaces the predicted spectrum with a non-geometric input would reveal how much of the 0.14 m accuracy is actually due to geometry versus transformer capacity.
  • If the LoS-plus-one-bounce assumption is the main limitation, the same architecture could be extended to include second-bounce paths or learned ray interactions, trading a modest increase in path enumeration for robustness in high-reflectivity rooms.
  • The ToF-offset augmentation could be generalized to other synchronization impairments, such as per-receiver clock skew or carrier frequency offset, using the same principle of sampling the impairment during training.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. The paper proposes GLocFM, a geometry-aware 3D indoor wireless localization model that takes WiFi channel observations and a 3D point cloud as inputs. The method formulates localization as maximum-likelihood estimation over a discrete candidate grid, uses a geometry-derived delay-AoA spectrum prior for LoS and one-bounce paths, and combines a cosine-similarity prior with a learned transformer-based scorer and an off-grid refinement head. A ToF-robust variant is trained with random time-offset augmentation. The model is trained on a new synthetic dataset of 221 Infinigen scenes with Sionna RT channels and evaluated on held-out synthetic scenes and on the real-measurement NeRF2 dataset. The reported mean 3D errors are 0.1429 m (NR=4) and 0.1746 m (NR=3) on synthetic data, and 0.0637 m on NeRF2, corresponding to 49.5%, 56.0%, and 48.8% reductions relative to the strongest LocGPT baseline.

Significance. If the reported results hold, GLocFM would be a meaningful advance: a single-forward-pass model that exploits scene geometry to localize transmitters across unseen scenes and real environments, with consistent improvements over several learning-based baselines. The paper's strengths include the construction of a multimodal synthetic dataset, the internal consistency of the headline numbers across Tables 1 and 2, the ablation of grid size, scorer, and path model in Table 3, the bandwidth/array robustness study, and a ToF-robust training scheme with a clear failure-mode analysis. The central claim is empirical, and several load-bearing details needed to verify it are currently underspecified or unquantified, notably per-seed variance, the exact NeRF2 adaptation, and the fidelity of the two-path geometric prior.

major comments (5)
  1. [Sec. 5.1.3, Tables 1-3] The text states that results are "obtained by averaging over multiple trials with different seeds," but no standard deviations, confidence intervals, or per-seed results are reported anywhere. This is load-bearing because Table 3 includes small differences such as the 0.7 cm gain of N=16 over N=12, and the robustness discussion in Sec. 5.2.1 relies on relative comparisons. Without a measure of spread, the reader cannot assess whether the headline improvements and ablation differences are statistically meaningful. Please report per-seed mean errors or error bars for all tables and figures, and specify the number of seeds.
  2. [Sec. 5.2.2, Table 2] The adaptation of GLocFM to the NeRF2 dataset is described only as applying "minor changes" to the predicted spectrum and the RF encoder inputs, because NeRF2 provides no OFDM CSI and no delay information. This adaptation is central to the real-data generalization claim, so the exact input representation, the modified predicted spectrum, and the fine-tuning protocol (including how the p% training subsets are chosen and whether the baselines receive the same adaptation) must be specified in enough detail to be reproducible.
  3. [Sec. 4.1.1, Eq. (14)] The predicted delay-AoA spectrum contains only LoS and one-bounce reflection terms, while the training data generated by Sionna include specular and diffuse reflections up to interaction depth three. The assertion that LoS and one-bounce paths convey the majority of the transmit power is not supported by any energy measurement in the paper. If second- and third-bounce or diffuse paths carry substantial energy in the evaluated scenes, the cosine prior in Eq. (15) compares the observed spectrum against a systematically incomplete template, and the learned scorer may compensate for the missing energy. This would weaken the paper's central attribution of cross-scene generalization to correct geometric priors. Please report the energy share of paths with two or more interactions and of diffuse paths in the synthetic dataset, or add an ablation that includes higher-order and diffuse terms in the predicted spectrum.
  4. [Sec. 4.1.3, Eqs. (8)-(10)] The quantities called "geometric priors" contain learned components: the path-loss scale beta_L, the soft-visibility attenuation gamma, and the calibration MLPs g_L and g_r are all trained with the same localization loss. The distinction between a physics-based geometric prior and a learned surrogate is therefore blurred. To support the claim that geometry, rather than learned compensation, drives the generalization, please clarify which parts are fixed physical models and which are learned, and consider ablating a fully fixed variant (e.g., free-space path loss with a binary visibility mask) to show how much of the gain relies on learned corrections.
  5. [Sec. 2.2 and Sec. 5.2.1] SigMap is identified in the related work as the method most closely related to GLocFM, since it also uses complex CFRs and treats a 3D map as a prompt in a foundation-model framework. However, SigMap is not included in either the synthetic benchmark (Table 1) or the real-data benchmark (Table 2). Given the paper's state-of-the-art claim, the closest related method should be benchmarked under the same setup, or the exclusion should be justified quantitatively.
minor comments (6)
  1. [Eq. (12)] The text says that a denominator normalizes the directional response by the received energy at the same delay bin, but no denominator appears in the displayed formula for C_r(v, l). Please correct the equation or the surrounding text.
  2. [Eq. (1)] The path gain alpha_{p,a} depends on the antenna element a, but the set P_r is defined only by the receiver index r. Please clarify whether the path set and its gains are per antenna element or per receiver with element-dependent phase/amplitude.
  3. [Sec. 5.1.3 and Table 3] The default grid is described as 12×12×4, but Table 3 reports only the horizontal grid size N. Please state whether N_z is held fixed in the grid-size ablation and whether the computational-complexity comparison includes the vertical dimension.
  4. [Sec. 5.3.2] The paragraph says that the model is fine-tuned for only a single epoch when applying it to different bandwidths and array sizes. Please clarify whether this is true for all points in Fig. 3(b), and whether longer fine-tuning changes the conclusions.
  5. [Throughout] There are several typographical errors and notation inconsistencies, including "arbitary," "gird," "developped," the rendering of square meters in Sec. 3, and the subscript placement in bFe(0.2). A careful proofread is needed.
  6. [Fig. 3(b)] The legend labels "80 MHz fixed" and "4×4 fixed" are ambiguous because the two panels are not clearly identified. Please state explicitly which configuration is held fixed in each panel.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central results are evaluated on held-out synthetic scenes and a real-measurement test split, and no claimed prediction reduces to a fitted input or to a self-citation.

full rationale

GLocFM's derivation chain is an end-to-end learned pipeline: observed CFRs are transformed into a delay-angular spectrum (Eqs. 11-12), candidate-dependent predicted spectra are synthesized from scene geometry plus learned attenuation terms (Eqs. 8-10 and 14), and a cosine similarity (Eq. 15) feeds a trained scoring function (Eqs. 16-18) whose parameters are optimized only on the training split by the L2 loss (Eq. 20). The paper never defines the predicted spectrum in terms of the observed spectrum or of the ground-truth position; the learned quantities beta_L, gamma, g_L, and g_r are fitted on training scenes and then frozen, and the reported errors are on 22 held-out synthetic scenes and on the NeRF2 real-measurement test split. The LoS/one-bounce assumption in Sec. 4.1.1 and the use of Sionna RT for the synthetic test set are modeling and domain concerns, not circular reductions: no equation equates the target output with a fitted parameter, and no load-bearing claim rests on a self-citation chain. The 'geometric prior' is learned rather than purely physical, which is an interpretation caveat, but it does not make the derivation circular by construction.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The claim is entirely empirical. The ledger records the learned components that make the predicted spectrum a fitted quantity, plus the domain assumptions required for synthetic-to-real transfer. There are no invented physical entities.

free parameters (6)
  • Learned path-loss scale beta_L = not reported in text
    Trained scalar in Eqs. (8) and (10); initialized to free-space path-loss exponent but optimized with the localization loss.
  • Soft-visibility attenuation gamma = not reported in text
    Learned positive factor in Eq. (7) controlling how opacity maps to visibility; part of the fitted geometric prior.
  • MLP calibration functions g_L and g_r = not reported, network weights
    Learned functions in Eqs. (8) and (9) that adjust LoS and reflection attenuation; make the predicted spectrum a fitted quantity.
  • Scene encoder, RF encoder, transformer, and scoring heads = trained on the synthetic GLocFM dataset
    All parameters are fit end-to-end with Eq. (20); the paper reports only aggregate loss and performance, not individual weights.
  • ToF augmentation standard deviation sigma_b = 4
    Hand-set in Section 5.4 for ToF-robust training; no procedure for choosing this value is given.
  • Model hyperparameters grid N, delay taps L, angular resolution = N=12, L=20, 5 degrees
    Defaults in Section 5.1.3; the ablation shows sensitivity to grid size, so these choices affect the headline numbers.
assumptions (6)
  • domain assumption LoS and one-bounce reflection paths carry the majority of transmitted power; higher-order and diffuse interactions can be neglected in the predicted spectrum.
    Stated in Section 4.1.1; the entire geometric prior is built on this simplification, while Sionna RT simulates up to depth three.
  • domain assumption Sionna RT, with EM properties assigned by semantic type, is a faithful simulator of indoor WiFi propagation for training a model that transfers to real measurements.
    Section 3; the synthetic training set and the apparent transfer to NeRF2 both depend on this.
  • domain assumption A surface-sampled point cloud is a sufficient geometric representation for predicting delay-AoA spectra.
    Section 3 converts meshes to point clouds and uses only those points for path synthesis and scene encoding.
  • domain assumption The NeRF2 real dataset provides geometry consistent with the environment of the recorded angular spectra.
    Section 5.2.2 evaluates GLocFM on NeRF2 without specifying how the 3D scene is obtained or aligned for the real measurements.
  • ad hoc to paper The learned scoring function is a valid surrogate for the intractable likelihood in Eq. (2).
    Section 4.2 replaces the likelihood with a trained scorer; no consistency or calibration guarantee is provided.
  • domain assumption Training on 177 synthetic rooms generalizes to unseen synthetic and real rooms.
    The whole method is empirical; no distributional argument is made, and the NeRF2 result is the only external check.

how reviews work

0 comments
Cite this review

Pith. "Pith review of GLocFM: A Geometry-Aware Foundation Model for 3D Indoor Wireless Localization." pith.science (2026). https://pith.science/paper/IN6YAQCH

@misc{pith2026260809285,
  author       = {Pith},
  title        = {Pith review of: GLocFM: A Geometry-Aware Foundation Model for 3D Indoor Wireless Localization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IN6YAQCH}},
  note         = {Machine review of arXiv:2608.09285}
}
abstract

Learning-based wireless localizers often fail to utilize geometric information about the propagation environment, limiting their ability to exploit non-line-of-sight (NLoS) propagation and generalize across scenes. To bridge this gap, we propose GLocFM, a Geometry-aware Localization Foundation Model, which jointly exploits WiFi measurements and scene geometry represented as a 3D point cloud. We formulate localization as a maximum-likelihood (ML) estimation problem, where the goal is to find a transmitter position that maximizes the likelihood of the wireless observations conditioned on the scene geometry. The likelihood of a candidate transmitter position is calculated by a learned scoring function that matches the observed delay--angle-of-arrival (AoA) spectrum against the spectrum predicted for that candidate. A hierarchical scene encoder extracts propagation-relevant features to produce geometric priors for LoS and one-bounce reflection paths. For scenarios with imperfect synchronization, we further introduce a time-of-flight (ToF)-robust GLocFM model to handle unknown ToF offsets. GLocFM is trained on a multi-modal synthetic indoor localization dataset comprising 221 diverse scenes whose associated wireless signals are generated using Sionna RT. On both synthetic and the NeRF$^{2}$ dataset based on real measurements, GLocFM reduces mean 3D localization error relative to one of the state-of-the-art localization baselines by 49.5\% and 48.8\%, respectively. Ablations across different number of receiver, bandwidths, and array sizes further demonstrate the effectiveness and robustness of the proposed framework.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

81 extracted references · 62 canonical work pages

  1. [1]

    IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews) , volume =

    Liu, Hui and Darabi, Houshang and Banerjee, Pat and Liu, Jing , title =. IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews) , volume =. 2007 , doi =

    2007

  2. [2]

    , title =

    Zafari, Faheem and Gkelias, Athanasios and Leung, Kin K. , title =. IEEE Communications Surveys & Tutorials , volume =. 2019 , doi =

    work page 2019

  3. [3]

    RadioDiff-3D: A 3D 3D Radio Map Dataset and Generative Diffusion Based Benchmark for 6G Environment-Aware Communication , year=

    Wang, Xiucheng and Zhang, Qiming and Cheng, Nan and Chen, Junting and Zhang, Zezhong and Li, Zan and Cui, Shuguang and Shen, Xuemin , journal=. RadioDiff-3D: A 3D 3D Radio Map Dataset and Generative Diffusion Based Benchmark for 6G Environment-Aware Communication , year=

    work page

  4. [4]

    2020 , eprint=

    Convolutional Occupancy Networks , author=. 2020 , eprint=

    work page 2020

  5. [5]

    International Conference on Indoor Positioning and Indoor Navigation , pages =

    Torres-Sospedra, Joaquin and Montoliu, Raul and Mart. International Conference on Indoor Positioning and Indoor Navigation , pages =. 2014 , doi =

    work page 2014

  6. [6]

    Proceedings of the 22nd Annual International Conference on Mobile Systems, Applications and Services , pages =

    Zhao, Xiaopeng and Wang, Guosheng and An, Zhenlin and Pan, Qingrui and Yang, Lei , title =. Proceedings of the 22nd Annual International Conference on Mobile Systems, Applications and Services , pages =. 2024 , doi =

    work page 2024

  7. [7]

    Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication , pages =

    Kotaru, Manikanta and Joshi, Kiran and Bharadia, Dinesh and Katti, Sachin , title =. Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication , pages =. 2015 , doi =

    work page 2015

  8. [8]

    Proceedings of the 18th Annual International Conference on Mobile Systems, Applications, and Services , pages =

    An, Zhenlin and Lin, Qiongzheng and Li, Ping and Yang, Lei , title =. Proceedings of the 18th Annual International Conference on Mobile Systems, Applications, and Services , pages =. 2020 , doi =

    work page 2020

Show all 81 references
  1. [9]

    Proceedings of the 26th Annual International Conference on Mobile Computing and Networking , pages =

    Ayyalasomayajula, Roshan and Arun, Aditya and Wu, Chenfeng and Sharma, Sanatan and Sethi, Abhishek Rajkumar and Vasisht, Deepak and Bharadia, Dinesh , title =. Proceedings of the 26th Annual International Conference on Mobile Computing and Networking , pages =. 2020 , doi =

    2020

  2. [10]

    2021 IEEE Global Communications Conference , pages =

    Vuckovic, Katarina and Hejazi, Farzam and Rahnavard, Nazanin , title =. 2021 IEEE Global Communications Conference , pages =. 2021 , doi =

    2021

  3. [11]

    , title =

    Kanhere, Ojas and Rappaport, Theodore S. , title =. IEEE Transactions on Wireless Communications , volume =. 2025 , doi =

    2025

  4. [12]

    Proceedings of the 29th Annual International Conference on Mobile Computing and Networking , pages =

    Zhao, Xiaopeng and An, Zhenlin and Pan, Qingrui and Yang, Lei , title =. Proceedings of the 29th Annual International Conference on Mobile Computing and Networking , pages =. 2023 , doi =

    2023

  5. [13]

    IEEE Transactions on Wireless Communications , volume =

    Levie, Ron and Yapar,. IEEE Transactions on Wireless Communications , volume =. 2021 , doi =

    2021

  6. [14]

    and Behboodi, Arash , title =

    Orekondy, Tribhuvanesh and Kumar, Pratik and Kadambi, Shreya and Ye, Hao and Soriaga, Joseph B. and Behboodi, Arash , title =. International Conference on Learning Representations , year =

  7. [15]

    International Conference on Learning Representations , year =

    Hehn, Thomas and Peschl, Markus and Orekondy, Tribhuvanesh and Behboodi, Arash and Brehmer, Johann , title =. International Conference on Learning Representations , year =

  8. [16]

    Proceedings of the 41st International Conference on Machine Learning , series =

    Lu, Haofan and Vattheuer, Christopher and Mirzasoleiman, Baharan and Abari, Omid , title =. Proceedings of the 41st International Conference on Machine Learning , series =. 2024 , url =

    2024

  9. [17]

    IEEE Journal on Multiscale and Multiphysics Computational Techniques , volume =

    Cao, Ge and Peng, Zhen , title =. IEEE Journal on Multiscale and Multiphysics Computational Techniques , volume =. 2024 , doi =

    2024

  10. [18]

    IEEE Transactions on Wireless Communications , volume =

    Zhang, Lihao and Sun, Haijian and Berweger, Samuel and Gentile, Camillo and Hu, Rose Qingyang , title =. IEEE Transactions on Wireless Communications , volume =. 2026 , doi =

    2026

  11. [19]

    2023 IEEE Globecom Workshops , pages =

    Hoydis, Jakob and Ait Aoudia, Faycal and Cammerer, Sebastian and Nimier-David, Merlin and Binder, Nikolaus and Marcus, Guillermo and Keller, Alexander , title =. 2023 IEEE Globecom Workshops , pages =. 2023 , doi =

    2023

  12. [20]

    2019 , eprint =

    Alkhateeb, Ahmed , title =. 2019 , eprint =

    2019

  13. [21]

    Study on Channel Model for Frequencies from 0.5 to 100 GHz , institution =

  14. [22]

    2025 , eprint =

    Han, Xueqiang and Zheng, Tianyue and Han, Tony Xiao and Luo, Jun , title =. 2025 , eprint =

    2025

  15. [23]

    , title =

    Qi, Charles Ruizhongtai and Yi, Li and Su, Hao and Guibas, Leonidas J. , title =. Advances in Neural Information Processing Systems , volume =

  16. [24]

    IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages =

    Wu, Xiaoyang and Jiang, Li and Wang, Peng-Shuai and Liu, Zhijian and Liu, Xihui and Qiao, Yu and Ouyang, Wanli and He, Tong and Zhao, Hengshuang , title =. IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages =. 2024 , doi =

    2024

  17. [25]

    Dataset of Pathloss and ToA Radio Maps With Localization Application , year =

    Yapar,. Dataset of Pathloss and ToA Radio Maps With Localization Application , year =. 2212.11777 , archivePrefix =

    arXiv

  18. [26]

    2026 , eprint =

    Qiao, Jing and Guo, Yiyang and Ye, Hao , title =. 2026 , eprint =

    2026

  19. [27]

    IEEE Transactions on Network Science and Engineering , volume =

    Wang, Xiucheng and Zhang, Qiming and Cheng, Nan and Chen, Junting and Zhang, Zezhong and Li, Zan and Cui, Shuguang and Shen, Xuemin , title =. IEEE Transactions on Network Science and Engineering , volume =. 2026 , doi =

    2026

  20. [28]

    ACM SIGGRAPH 2026 Conference Papers , year =

    Xu, Bing and Varma T, Mukund and Wang, Cheng and Li, Tzu-Mao and Wu, Lifan and Wronski, Bartlomiej and Ramamoorthi, Ravi and Salvi, Marco , title =. ACM SIGGRAPH 2026 Conference Papers , year =

    2026

  21. [29]

    IEEE Transactions on Wireless Communications , volume =

    Wen, Chaozheng and Tong, Jingwen and Hu, Yingdong and Lin, Zehong and Zhang, Jun , title =. IEEE Transactions on Wireless Communications , volume =. 2026 , doi =

    2026

  22. [30]

    2026 , eprint =

    Wen, Chaozheng and Tong, Jingwen and Lin, Zehong and Bian, Chenghong and Zhang, Jun , title =. 2026 , eprint =

    2026

  23. [31]

    Lee , title =

    Yang, Xi and Wen, Chao-Kai and Han, Yu and Jin, Shi and Swindlehurst, A. Lee , title =. IEEE Transactions on Signal Processing , volume =. 2022 , doi =

    2022

  24. [32]

    IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , pages =

    Yapar,. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , pages =. 2022 , doi =

    2022

  25. [33]

    IEEE Transactions on Vehicular Technology , volume =

    Keum, Inkook and Son, Jinwoo and Kim, Hyunsoo and Moon, Jihoon and Shim, Byonghyo , title =. IEEE Transactions on Vehicular Technology , volume =. 2026 , doi =

    2026

  26. [34]

    The Fourteenth International Conference on Learning Representations , year =

    Chu, Yong and Zhou, Xun and Xu, Zenglin and Wang, Hui and Yu, Yue , title =. The Fourteenth International Conference on Learning Representations , year =

  27. [35]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages =

    Raistrick, Alexander and Mei, Lingjie and Kayan, Karhan and Yan, David and Zuo, Yiming and Han, Beining and Wen, Hongyu and Parakh, Meenal and Alexandropoulos, Stamatis and Lipson, Lahav and Ma, Zeyu and Deng, Jia , title =. Proceedings of the IEEE/CVF Conference on Computer V...

  28. [36]

    Joint Device Positioning and Clock Synchronization in 5G Ultra-Dense Networks , journal =

    Koivisto, Mike and Costa, M. Joint Device Positioning and Clock Synchronization in 5G Ultra-Dense Networks , journal =. 2017 , doi =

    2017

  29. [37]

    10th USENIX Symposium on Networked Systems Design and Implementation (NSDI 13) , pages =

    Xiong, Jie and Jamieson, Kyle , title =. 10th USENIX Symposium on Networked Systems Design and Implementation (NSDI 13) , pages =. 2013 , publisher =

    2013

  30. [38]

    13th USENIX Symposium on Networked Systems Design and Implementation (NSDI 16) , pages =

    Vasisht, Deepak and Kumar, Swarun and Katabi, Dina , title =. 13th USENIX Symposium on Networked Systems Design and Implementation (NSDI 16) , pages =. 2016 , publisher =

    2016

  31. [39]

    Proceedings of the 21st Annual International Conference on Mobile Computing and Networking , pages =

    Xiong, Jie and Sundaresan, Karthikeyan and Jamieson, Kyle , title =. Proceedings of the 21st Annual International Conference on Mobile Computing and Networking , pages =. 2015 , publisher =

    2015

  32. [40]

    Proceedings of the 25th Annual International Conference on Mobile Computing and Networking , pages =

    Xie, Yaxiong and Xiong, Jie and Li, Mo and Jamieson, Kyle , title =. Proceedings of the 25th Annual International Conference on Mobile Computing and Networking , pages =. 2019 , publisher =

    2019

  33. [41]

    2015 IEEE Wireless Communications and Networking Conference (WCNC) , pages =

    Wang, Xuyu and Gao, Lingjun and Mao, Shiwen and Pandey, Santosh , title =. 2015 IEEE Wireless Communications and Networking Conference (WCNC) , pages =. 2015 , publisher =

    2015

  34. [42]

    IEEE Internet of Things Journal , volume =

    Wang, Xuyu and Gao, Lingjun and Mao, Shiwen , title =. IEEE Internet of Things Journal , volume =. 2016 , doi =

    2016

  35. [43]

    IEEE Access , volume =

    Chen, Hao and Zhang, Yifan and Li, Wei and Tao, Xiaofeng and Zhang, Ping , title =. IEEE Access , volume =. 2017 , doi =

    2017

  36. [44]

    IEEE Access , volume =

    Zhang, Zhongfeng and Du, Hongxin and Choi, Seungwon and Cho, Sung Ho , title =. IEEE Access , volume =. 2022 , doi =

    2022

  37. [45]

    Channel Charting: Locating Users Within the Radio Environment Using Channel State Information , journal =

    Studer, Christoph and Medjkouh, Sa. Channel Charting: Locating Users Within the Radio Environment Using Channel State Information , journal =. 2018 , doi =

    2018

  38. [46]

    IEEE Transactions on Wireless Communications , volume =

    Gentner, Christian and Jost, Thomas and Wang, Wei and Zhang, Siwei and Dammann, Armin and Fiebig, Uwe-Carsten , title =. IEEE Transactions on Wireless Communications , volume =. 2016 , doi =

    2016

  39. [47]

    17th USENIX Symposium on Networked Systems Design and Implementation (NSDI 20) , pages =

    Ayyalasomayajula, Roshan and Arun, Aditya and Wu, Chenfeng and Rajagopalan, Shrivatsan and Ganesaraman, Shreya and Seetharaman, Aravind and Jain, Ish Kumar and Bharadia, Dinesh , title =. 17th USENIX Symposium on Networked Systems Design and Implementation (NSDI 20) , pages =....

    2020

  40. [48]

    IEEE Access , volume =

    Suga, Norisato and Maeda, Yoshihiro and Sato, Koya , title =. IEEE Access , volume =. 2023 , doi =

    2023

  41. [49]

    Deepmimo: A generic deep learning dataset for millimeter wave and massive mimo applications, 2019

    Ahmed Alkhateeb. Deepmimo: A generic deep learning dataset for millimeter wave and massive mimo applications, 2019

    2019

  42. [50]

    General-purpose deep tracking platform across protocols for the internet of things

    Zhenlin An, Qiongzheng Lin, Ping Li, and Lei Yang. General-purpose deep tracking platform across protocols for the internet of things. In Proceedings of the 18th Annual International Conference on Mobile Systems, Applications, and Services, pages 94--106, 2020. doi:10.1145/338...

    2020

  43. [51]

    LocAP : Autonomous millimeter accurate mapping of WiFi infrastructure

    Roshan Ayyalasomayajula, Aditya Arun, Chenfeng Wu, Shrivatsan Rajagopalan, Shreya Ganesaraman, Aravind Seetharaman, Ish Kumar Jain, and Dinesh Bharadia. LocAP : Autonomous millimeter accurate mapping of WiFi infrastructure. In 17th USENIX Symposium on Networked Systems Design ...

    2020

  44. [52]

    Deep learning based wireless localization for indoor navigation

    Roshan Ayyalasomayajula, Aditya Arun, Chenfeng Wu, Sanatan Sharma, Abhishek Rajkumar Sethi, Deepak Vasisht, and Dinesh Bharadia. Deep learning based wireless localization for indoor navigation. In Proceedings of the 26th Annual International Conference on Mobile Computing and ...

    2020

  45. [53]

    ConFi : Convolutional neural networks based indoor Wi-Fi localization using channel state information

    Hao Chen, Yifan Zhang, Wei Li, Xiaofeng Tao, and Ping Zhang. ConFi : Convolutional neural networks based indoor Wi-Fi localization using channel state information. IEEE Access, 5: 0 18066--18074, 2017. doi:10.1109/ACCESS.2017.2749516

    2017

  46. [54]

    Map as a prompt: Learning multi-modal spatial-signal foundation models for cross-scenario wireless localization

    Yong Chu, Xun Zhou, Zenglin Xu, Hui Wang, and Yue Yu. Map as a prompt: Learning multi-modal spatial-signal foundation models for cross-scenario wireless localization. In The Fourteenth International Conference on Learning Representations, 2026. URL https://openreview.net/forum...

    2026

  47. [55]

    Multipath assisted positioning with simultaneous localization and mapping

    Christian Gentner, Thomas Jost, Wei Wang, Siwei Zhang, Armin Dammann, and Uwe-Carsten Fiebig. Multipath assisted positioning with simultaneous localization and mapping. IEEE Transactions on Wireless Communications, 15 0 (9): 0 6104--6117, 2016. doi:10.1109/TWC.2016.2578336

    2016

  48. [56]

    Rayloc: Wireless indoor localization via fully differentiable ray-tracing, 2025

    Xueqiang Han, Tianyue Zheng, Tony Xiao Han, and Jun Luo. Rayloc: Wireless indoor localization via fully differentiable ray-tracing, 2025

    2025

  49. [57]

    Differentiable and learnable wireless simulation with geometric transformers

    Thomas Hehn, Markus Peschl, Tribhuvanesh Orekondy, Arash Behboodi, and Johann Brehmer. Differentiable and learnable wireless simulation with geometric transformers. In International Conference on Learning Representations, 2025. URL https://openreview.net/forum?id=9TClCDZXeh

    2025

  50. [58]

    Sionna rt: Differentiable ray tracing for radio propagation modeling

    Jakob Hoydis, Faycal Ait Aoudia, Sebastian Cammerer, Merlin Nimier-David, Nikolaus Binder, Guillermo Marcus, and Alexander Keller. Sionna rt: Differentiable ray tracing for radio propagation modeling. In 2023 IEEE Globecom Workshops, pages 317--321, 2023. doi:10.1109/GCWkshps5...

    2023

  51. [59]

    Rappaport

    Ojas Kanhere and Theodore S. Rappaport. Map-assisted millimeter wave and terahertz position location and sensing. IEEE Transactions on Wireless Communications, 24 0 (6): 0 5323--5336, 2025. doi:10.1109/TWC.2025.3546746

    2025

  52. [60]

    Deep learning-based NLoS localization using geometric map image

    Inkook Keum, Jinwoo Son, Hyunsoo Kim, Jihoon Moon, and Byonghyo Shim. Deep learning-based NLoS localization using geometric map image. IEEE Transactions on Vehicular Technology, 75 0 (6): 0 11758--11763, 2026. doi:10.1109/TVT.2025.3649253

    2026

  53. [61]

    Joint device positioning and clock synchronization in 5g ultra-dense networks

    Mike Koivisto, M \'a rio Costa, Janis Werner, Kari Heiska, Jukka Talvitie, Kari Lepp \"a nen, Visa Koivunen, and Mikko Valkama. Joint device positioning and clock synchronization in 5g ultra-dense networks. IEEE Transactions on Wireless Communications, 16 0 (5): 0 2866--2881, ...

    2017

  54. [62]

    Spotfi: Decimeter level localization using wifi

    Manikanta Kotaru, Kiran Joshi, Dinesh Bharadia, and Sachin Katti. Spotfi: Decimeter level localization using wifi. In Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication, pages 269--282, 2015. doi:10.1145/2785956.2787487

    2015

  55. [63]

    Survey of wireless indoor positioning techniques and systems

    Hui Liu, Houshang Darabi, Pat Banerjee, and Jing Liu. Survey of wireless indoor positioning techniques and systems. IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), 37 0 (6): 0 1067--1080, 2007. doi:10.1109/TSMCC.2007.905750

    2007

  56. [64]

    Convolutional occupancy networks, 2020

    Songyou Peng, Michael Niemeyer, Lars Mescheder, Marc Pollefeys, and Andreas Geiger. Convolutional occupancy networks, 2020. URL https://arxiv.org/abs/2003.04618

    2020 arXiv

  57. [65]

    Infinigen indoors: Photorealistic indoor scenes using procedural generation

    Alexander Raistrick, Lingjie Mei, Karhan Kayan, David Yan, Yiming Zuo, Beining Han, Hongyu Wen, Meenal Parakh, Stamatis Alexandropoulos, Lahav Lipson, Zeyu Ma, and Jia Deng. Infinigen indoors: Photorealistic indoor scenes using procedural generation. In Proceedings of the IEEE...

    2024

  58. [66]

    i d Medjkouh, Emre G \

    Christoph Studer, Sa \"i d Medjkouh, Emre G \"o n \"u lta s , Tom Goldstein, and Olav Tirkkonen. Channel charting: Locating users within the radio environment using channel state information. IEEE Access, 6: 0 47682--47698, 2018. doi:10.1109/ACCESS.2018.2866979

    2018

  59. [67]

    Indoor radio map construction via ray tracing with RGB-D sensor-based 3d reconstruction: Concept and experiments in WLAN systems

    Norisato Suga, Yoshihiro Maeda, and Koya Sato. Indoor radio map construction via ray tracing with RGB-D sensor-based 3d reconstruction: Concept and experiments in WLAN systems. IEEE Access, 11: 0 24863--24874, 2023. doi:10.1109/ACCESS.2023.3254912

    2023

  60. [68]

    Avariento, Tom \'a s J

    Joaquin Torres-Sospedra, Raul Montoliu, Adolfo Mart \'i nez-Us \'o , Joan P. Avariento, Tom \'a s J. Arnau, Mauri Benedito-Bordonau, and Joaqu \'i n Huerta. UJIIndoorLoc : A new multi-building and multi-floor database for WLAN fingerprint-based indoor localization problems. In...

    2014

  61. [69]

    Decimeter-level localization with a single WiFi access point

    Deepak Vasisht, Swarun Kumar, and Dina Katabi. Decimeter-level localization with a single WiFi access point. In 13th USENIX Symposium on Networked Systems Design and Implementation (NSDI 16), pages 165--178. USENIX Association, 2016. URL https://www.usenix.org/conference/nsdi1...

    2016

  62. [70]

    Map-csi: Single-site map-assisted localization using massive mimo csi

    Katarina Vuckovic, Farzam Hejazi, and Nazanin Rahnavard. Map-csi: Single-site map-assisted localization using massive mimo csi. In 2021 IEEE Global Communications Conference, pages 1--6, 2021. doi:10.1109/GLOBECOM46510.2021.9685564

    2021

  63. [71]

    DeepFi : Deep learning for indoor fingerprinting using channel state information

    Xuyu Wang, Lingjun Gao, Shiwen Mao, and Santosh Pandey. DeepFi : Deep learning for indoor fingerprinting using channel state information. In 2015 IEEE Wireless Communications and Networking Conference (WCNC), pages 1666--1671. IEEE, 2015. doi:10.1109/WCNC.2015.7127718

    2015

  64. [72]

    CSI phase fingerprinting for indoor localization with a deep learning approach

    Xuyu Wang, Lingjun Gao, and Shiwen Mao. CSI phase fingerprinting for indoor localization with a deep learning approach. IEEE Internet of Things Journal, 3 0 (6): 0 1113--1123, 2016. doi:10.1109/JIOT.2016.2558659

    2016

  65. [73]

    mD-Track : Leveraging multi-dimensionality in passive indoor Wi-Fi tracking

    Yaxiong Xie, Jie Xiong, Mo Li, and Kyle Jamieson. mD-Track : Leveraging multi-dimensionality in passive indoor Wi-Fi tracking. In Proceedings of the 25th Annual International Conference on Mobile Computing and Networking, pages 1--16. ACM, 2019. doi:10.1145/3300061.3300133

    2019

  66. [74]

    ArrayTrack : A fine-grained indoor location system

    Jie Xiong and Kyle Jamieson. ArrayTrack : A fine-grained indoor location system. In 10th USENIX Symposium on Networked Systems Design and Implementation (NSDI 13), pages 71--84. USENIX Association, 2013. URL https://www.usenix.org/conference/nsdi13/technical-sessions/presentat...

    2013

  67. [75]

    ToneTrack : Leveraging frequency-agile radios for time-based indoor wireless localization

    Jie Xiong, Karthikeyan Sundaresan, and Kyle Jamieson. ToneTrack : Leveraging frequency-agile radios for time-based indoor wireless localization. In Proceedings of the 21st Annual International Conference on Mobile Computing and Networking, pages 537--549. ACM, 2015. doi:10.114...

    2015

  68. [76]

    Lee Swindlehurst

    Xi Yang, Chao-Kai Wen, Yu Han, Shi Jin, and A. Lee Swindlehurst. Soft channel estimation and localization for millimeter wave systems with multiple receivers. IEEE Transactions on Signal Processing, 70: 0 4897--4911, 2022. doi:10.1109/TSP.2022.3192584

    2022

  69. [77]

    LocUNet : Fast urban positioning using radio maps and deep learning

    C a g kan Yapar, Ron Levie, Gitta Kutyniok, and Giuseppe Caire. LocUNet : Fast urban positioning using radio maps and deep learning. In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 4063--4067, 2022. doi:10.1109/ICASSP43922.2022.9747240

    2022

  70. [78]

    Faheem Zafari, Athanasios Gkelias, and Kin K. Leung. A survey of indoor localization systems and technologies. IEEE Communications Surveys & Tutorials, 21 0 (3): 0 2568--2599, 2019. doi:10.1109/COMST.2019.2911558

    2019

  71. [79]

    TIPS : Transformer based indoor positioning system using both CSI and DoA of WiFi signal

    Zhongfeng Zhang, Hongxin Du, Seungwon Choi, and Sung Ho Cho. TIPS : Transformer based indoor positioning system using both CSI and DoA of WiFi signal. IEEE Access, 10: 0 111363--111376, 2022. doi:10.1109/ACCESS.2022.3215504

    2022

  72. [80]

    Nerf ^ 2 : Neural radio-frequency radiance fields

    Xiaopeng Zhao, Zhenlin An, Qingrui Pan, and Lei Yang. Nerf ^ 2 : Neural radio-frequency radiance fields. In Proceedings of the 29th Annual International Conference on Mobile Computing and Networking, pages 1--15, 2023. doi:10.1145/3570361.3592527

    2023

  73. [81]

    Understanding localization by a tailored gpt

    Xiaopeng Zhao, Guosheng Wang, Zhenlin An, Qingrui Pan, and Lei Yang. Understanding localization by a tailored gpt. In Proceedings of the 22nd Annual International Conference on Mobile Systems, Applications and Services, pages 318--330, 2024. doi:10.1145/3643832.3661869

    2024

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.