OWP-IMU: An RSS-based Optical Wireless and IMU Indoor Positioning Dataset

Daan Delabie; Fan Wu; Francois Rottenberg; Jorik De Bruycker; Lieven De Strycker; Nobby Stevens

REVIEW 4 major objections 5 minor 25 references

OWP-IMU: An RSS-based Optical Wireless and IMU Indoor Positioning Dataset

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read The paper introduces the first open-source dataset fusing optical-wireless RSS, IMU, and motion-capture ground truth for indoor positioning.

desk verdict A genuinely useful open dataset for RSS optical wireless positioning plus IMU; the benchmark claims need small but real fixes, and the timestamp synchronization must be documented. read the letter →

arxiv 2505.16823 v1 pith:LC4TLJTF submitted 2025-05-22 eess.SP

Fan Wu , Jorik De Bruycker , Daan Delabie , Nobby Stevens , Francois Rottenberg , Lieven De Strycker This is my paper · ORCID

classification eess.SP

keywords receivedsignalstrengthopticalwirelesspositioningvisiblelightindoorlocalizationinertialmeasurementunitextendedKalmanfilterGaussianprocessregressionNLOSbenchmark

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's objective is to give the indoor-localization community a public, synchronized dataset for received-signal-strength based optical wireless positioning combined with inertial data. The authors recorded infrared RSS at 27 Hz, IMU readings at 200 Hz, and ground-truth position and orientation at 160 Hz while a small vehicle drove continuous trajectories in two rooms, one obstacle-free and one with a metal column blocking line of sight, at three speeds. The resulting OWP-IMU dataset holds over 110,000 samples and more than 80 minutes of motion. A sympathetic reading of the paper is that these characteristics make it the first open benchmark of its kind, giving researchers a common testbed to develop and compare OWP and OWP-plus-IMU localization algorithms. The authors also show the dataset is usable by benchmarking Gaussian-process regression and extended-Kalman-filter fusion.

What carries the argument

The load-bearing object is the synchronized triple-stream recording: OWP RSS samples at 27 Hz, IMU readings at 200 Hz, and motion-capture ground truth at 160 Hz, collected on a vehicle with four ceiling-mounted infrared LEDs transmitting at distinct frequencies. The paper uses Gaussian-process regression to map the four RSS values to 2D position, multilateration with a Lambertian channel model as a classical baseline, and an extended Kalman filter whose state includes position, velocity, heading, accelerations, and accelerometer biases, with zero-velocity updates during the stationary segment at the start of each recording. This machinery lets the paper demonstrate both the dataset's standalone usability and the accuracy gain from fusing inertial data.

What would settle it

An independent audit could take the released files and re-synchronize the streams by estimating clock offset and drift instead of taking the nearest ground-truth timestamp for each OWP sample; if the P99 errors move by more than a few centimeters, the reported 40 cm-to-25 cm improvement is partly an artifact of the alignment assumption. A second check would include the 19 mm IMU-to-photodiode lever arm in the frame alignment and recompute the EKF errors.

Watch

Extended reading notes

Core claim

The central claim is that OWP-IMU is the first open-source dataset pairing received-signal-strength optical wireless positioning with IMU data and accurate ground truth, containing continuous trajectories at three velocities under both line-of-sight and non-line-of-sight conditions. The dataset records four infrared LED RSS values at 27 Hz, IMU acceleration and gyroscope readings at 200 Hz, and motion-capture position and orientation at 160 Hz, synchronized so each OWP sample can be matched to a ground-truth position. The benchmark results show that Gaussian-process regression on the RSS alone reaches roughly 10 cm median error with 400 training samples, while an extended Kalman filter combining IMU and OWP reduces the P99 error from about 40 cm to about 25 cm and tracks heading with about 8 degrees mean absolute error over more than 10 minutes of motion.

Load-bearing premise

The benchmark numbers assume that matching each optical sample to the nearest ground-truth time stamp gives the true position and that the IMU and motion-capture frames are aligned with the 19 mm IMU-to-photodiode offset being negligible.

Editorial extensions

If this is right

Researchers can train and evaluate RSS-based optical wireless positioning methods on open data without building their own optical transmitter and receiver testbed.
The 27 Hz RSS and 200 Hz IMU rates with synchronized ground truth give a concrete benchmark for data-efficient learning and sensor-fusion algorithms.
GP regression reaching about 10 cm median error with 400 samples implies that small training sets suffice for fingerprint-style positioning in this environment.
The EKF results imply that adding IMU data mainly suppresses outlier errors, cutting P99 from about 40 cm to about 25 cm while keeping median error near 8 to 10 cm.
The three speeds and the obstacle versus no-obstacle contrast provide controlled conditions for studying how dynamics and non-line-of-sight conditions affect localization error.

Reading between the lines

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

A natural extension the authors do not pursue is using the one-minute stationary segment at the start of each recording to calibrate IMU bias per trajectory, which would test how much of the EKF gain relies on the built-in zero-velocity reset.
Because the 27 Hz RSS stream and 200 Hz IMU stream are synchronized to a common ground truth, the dataset could also support learning-based fusion methods that predict position between optical updates, a direction the paper only mentions as future work.
If the timestamp-matching assumption is violated by clock drift, the reported error numbers would shift; re-estimating the clock model from the data would reveal how much of the 40 cm to 25 cm P99 improvement is alignment-dependent.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Plain take: the paper gives the field something it didn't have: an open, synchronized RSS optical wireless plus IMU dataset with continuous trajectories, orientation ground truth, three speeds, and a real NLOS obstruction. That is a useful artifact and it should become a reference for benchmarking indoor OWP fusion.

What's new and good: Table I is honest about prior work. Compared to the 158 static points in [16] and the 98 camera AOA points in [17], this is a step change in scale and scope. The platform description is careful: LED positions, PD height, IMU-to-PD offset, motion-capture GT, stationary intervals for ZUPT. The GP and EKF benchmarks are standard, but they demonstrate the dataset is usable and give future users a baseline.

Soft spots, in order. The headline "45% reduction" in P99 is an arithmetic overclaim: 40 cm to 25 cm is 37.5%. Fix the number. Multilateration gains are calibrated on all measurement datasets and then used to evaluate multilateration on the same data; that is in-sample and should be cross-validated. The gap between "over 110 k" OWP samples and "about 17 000" synced samples is unexplained and needs a sentence. And the benchmarks report percentiles but no variance or error bars; ten random splits should give you spread.

The timestamp synchronization question is the one I'd push hardest. The paper says only that for each OWP timestamp you find the closest GT timestamp and take that as ground truth. No shared clock, no drift compensation, no validation of the alignment. If clock drift or latency is present, every paired position is wrong in a way that is hard to see in aggregate. You don't strictly need to fix it for the dataset to be useful, but you do need to document how the streams were synchronized and provide some check, for example comparing stationary periods. The stress-test note is right that this is the load-bearing assumption, and it is currently underexplained. The 19 mm IMU-to-PD offset is much less concerning; it is small and fixed.

Indoor positioning and sensor fusion researchers will want this dataset. The paper deserves serious peer review, with the benchmark cleanup and the synchronization section as required revisions. I would accept it as a dataset paper rather than desk reject.

Referee Report

4 major / 5 minor

Summary. This paper presents OWP-IMU, an open dataset of received-signal-strength optical wireless positioning (OWP) measurements, IMU data, and Qualisys motion-capture ground truth collected in two indoor environments (clear and with a metal-column obstacle) at three vehicle speeds. The authors describe the hardware platform, data formats, and two benchmark families: Gaussian-process regression and multilateration on RSS alone, and an extended Kalman filter fusing IMU and RSS. The central claim is that this is the first open-source dataset combining RSS-based OWP, IMU, and accurate ground truth with continuous trajectories, LOS/NLOS conditions, three velocities, over 110k samples, and more than 80 minutes of recording.

Significance. If the central claim holds, OWP-IMU would be a valuable community testbed: it offers high-rate OWP (27 Hz), IMU (200 Hz), and ground truth (160 Hz) with continuous trajectories and controlled LOS/NLOS conditions, and the paper includes benchmark numbers that let users anticipate performance. Strengths include a detailed mechanical platform description, use of an independent motion-capture system for ground truth, a public dataset link, and reproducible baseline methods. The main limitations are the undocumented timestamp synchronization between streams and the in-sample calibration of the multilateration benchmark, both of which must be addressed before the dataset's synchronization claim and benchmark comparisons can be fully trusted.

major comments (4)

[Section IV.A] The paper does not describe how the OWP, IMU, and ground-truth timestamp streams are synchronized. The only statement is 'for each OWP timestamp we find the closest GT timestamp and take its coordinates as the true position for that sample.' This assumes a common clock or a known, drift-free offset between devices. If the OWP and Qualisys clocks have relative drift or a constant latency, nearest-neighbor matching will systematically pair each RSS sample to the wrong ground-truth position, and every downstream benchmark and any user of the dataset will inherit that error. Please document the synchronization method (shared clock, PTP, post-hoc alignment) and provide a validation experiment, for example a static recording showing the residual synchronization error.
[Section IV.A / Table I] The manuscript claims 'over 110 k data points' and 'Receiver Total Captures > 110 k' (Table I), yet Section IV.A says 'The total number of synced datasets is about 17000.' At 27 Hz OWP sampling over more than 80 min one would expect roughly 130,000 OWP samples, so the two numbers cannot both describe the same set. Please clarify whether 110k is the raw OWP count, a sum over all three streams, or something else, and report per-trajectory raw and synchronized sample counts, including how the stationary initial minute and any dropped samples are treated. Without this breakdown, the dataset size and the 17k-sample benchmark cannot be interpreted.
[Section IV.A] The multilateration benchmark is calibrated in-sample: 'We first calibrate a gain factor for each LED by all measurement datasets. Then we use it to compute each LED to PD distance from its RSS reading' and the same datasets are used for the reported errors. This gives multilateration an advantage that the Gaussian-process benchmark, which is evaluated on held-out splits, does not receive, and it makes the multilateration-vs-GP comparison and the reported multilateration errors unrepresentative of unseen conditions. Please calibrate the gain factors on a training subset or with leave-one-trajectory-out validation and report calibration and test errors separately.
[Section IV.B] The EKF benchmark description does not state whether the 400-sample GP model used inside the EKF is trained on data from the same trajectory being evaluated or on separate training data. If the same trajectory is used for both training and evaluation, the position updates inside the EKF are trained in-sample, which can only improve the reported P50/P99 numbers. Please specify the training/evaluation split for the GP model used in the EKF and, ideally, report leave-one-trajectory-out results.

minor comments (5)

[Table I] Table I marks reference [16] as not open access despite it being published on IEEE DataPort with a DOI; please verify and correct the table.
[Section II / Figure 3] The 19 mm vertical offset between the IMU and the PD is mentioned but its effect on the EKF's position and heading estimates is not quantified; a short sensitivity note would help users.
[Section IV.B] The EKF implementation is not reproducible without the process/measurement noise covariance matrices and the GP kernel hyperparameters; please include these values or provide the code in the dataset release.
[Section IV / Figure 7] Figure 7 attributes a large OWP error to 'shadows or vehicle vibration' without evidence; a timing-misalignment artifact could also produce such outliers, so this interpretation should be revisited once synchronization is validated.
[Section III / Table III] Minor wording: 'The total number of synced datasets is about 17000' should read 'synchronized OWP samples'; also clarify whether the GT rotation matrix is stored row-major or column-major.

Circularity Check

1 steps flagged · score 4.0 of 10

Dataset itself is externally grounded; the only circular step is the in-sample multilateration gain calibration in the benchmark.

fitted input called prediction [Section IV.A (OWP Benchmarks)]
"We first calibrate a gain factor for each LED by all measurement datasets. Then we use it to compute each LED to PD distance from its RSS reading. Finally, we solve for the receiver position by a least-squares intersection of the four distance estimates."

The multilateration gain factor is fitted on all measurement datasets and then used to produce distance estimates and position errors on those same datasets. The P50/P99 multilateration results in Table IV and Figure 6 are therefore in-sample fits, not out-of-sample predictions. The reported error is partly forced by the fitted gain rather than measuring generalization, and comparing this to GP results from a train/test split is an uneven benchmark. This circularity is confined to the benchmark; the central dataset claim is grounded in external Qualisys ground truth and does not reduce to the fitted parameters.

full rationale

The paper's central contribution is a measured dataset: RSS, IMU, and ground-truth positions from an external Qualisys motion-capture system. That dataset is not derived from any model, and the 'first open-source dataset' claim is checked against cited prior datasets [16] and [17], so there is no definitional circularity or load-bearing self-citation. The one genuinely circular element is the multilateration benchmark in Section IV.A: a gain factor per LED is calibrated on all measurement datasets and then used to evaluate multilateration on the same datasets, making the reported errors in-sample rather than predictive. This does not undermine the dataset itself, but it does mean the multilateration benchmark numbers are optimistically biased and not comparable to the GP train/test results. The EKF benchmark also relies on a GP model trained on the same recordings, which is a weaker form of in-sample evaluation, but it is not as cleanly reducible to a fitted parameter. Timestamp alignment is an important correctness assumption but is not a circularity. Overall, the central claim retains independent content, so the appropriate score is 4, reflecting one partial circularity in a benchmark rather than in the core dataset contribution.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

No new physical entities are introduced. The paper's contribution is a measured dataset and benchmark evaluations; the only fitted quantities are benchmark calibration and tuning parameters, which are listed above.

free parameters (3)

Per-LED gain factor (multilateration) = not reported
A gain factor for each of the four LEDs is calibrated over all measurement datasets and used to convert RSS to distance for evaluation on the same datasets; these are fitted parameters.
Gaussian process kernel hyperparameters = not reported
GP regression requires kernel and noise hyperparameters trained on data; the paper does not state kernel choice, lengthscales, or noise.
EKF process and measurement noise covariances = not reported
The extended Kalman filter needs noise covariance matrices and an initial state; these tuning values are not given in the paper.

assumptions (4)

domain assumption The optical channel follows RSS proportional to 1/d^(m+3) with Lambertian order m=0.5 and no transmitter or receiver tilt.
Invoked in Section IV.A for multilateration distance estimation; if tilt or channel exponent differ, distance estimates and resulting localization errors are biased.
domain assumption The IMU frame is parallel to the motion capture frame and the 19 mm offset between IMU and photodiode is negligible.
Stated in Sections II and III; the EKF fuses IMU measurements assuming this alignment, and any residual misalignment biases acceleration and heading estimates.
domain assumption A random subset of trajectory points can be used for training a GP and the remaining points as an unbiased test set.
Used in Section IV.A; trajectory samples are strongly correlated in time and space, so random splits may overstate generalization accuracy.
domain assumption Qualisys motion capture output is treated as ground truth without reported error bars.
The paper uses motion capture positions and orientations as true values throughout the benchmarks; any motion capture noise propagates into all reported errors.

how reviews work

0 comments

Cite this review

Pith. "Pith review of OWP-IMU: An RSS-based Optical Wireless and IMU Indoor Positioning Dataset." pith.science (2026). https://pith.science/paper/LC4TLJTF

@misc{pith2026250516823,
  author       = {Pith},
  title        = {Pith review of: OWP-IMU: An RSS-based Optical Wireless and IMU Indoor Positioning Dataset},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LC4TLJTF}},
  note         = {Machine review of arXiv:2505.16823}
}

read the original abstract

Received signal strength (RSS)-based optical wireless positioning (OWP) systems are becoming popular for indoor localization because they are low-cost and accurate. However, few open-source datasets are available to test and analyze RSS-based OWP systems. In this paper, we collected RSS values at a sampling frequency of 27 Hz, inertial measurement unit (IMU) at a sampling frequency of 200 Hz and the ground truth at a sampling frequency of 160 Hz in two indoor environments. One environment has no obstacles, and the other has a metal column as an obstacle to represent a non-line-of-sight (NLOS) scenario. We recorded data with a vehicle at three different speeds (low, medium and high). The dataset includes over 110 k data points and covers more than 80 min. We also provide benchmark tests to show localization performance using only RSS-based OWP and improve accuracy by combining IMU data via extended kalman filter. The dataset OWP-IMU is open source1 to support further research on indoor localization methods.

Figures

Figures reproduced from arXiv: 2505.16823 by the authors.

**Figure 1.** Illustration of the experimental setup for data coll [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗

**Figure 3.** Top and side views of our customized platform com [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗

Figures from the paper (4 more)

**Figure 4.** Figure 4: Trajectories recorded by the vehicle at different sp [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]

**Figure 5.** Figure 5: IMU acceleration measurement example under low [PITH_FULL_IMAGE:figures/full_fig_p004_5.png]

**Figure 6.** Figure 6: Error distributions of GPs training models at three v [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]

**Figure 7.** Figure 7: The GT, EKF and OWP tracked trajectory with medium [PITH_FULL_IMAGE:figures/full_fig_p005_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 25 canonical work pages

[16]

Machine learning for rss-based visible light positioning ,

W. Raes, J. De Bruycker, N. Knudde, T. Dhaene, and N. Stev ens, “Machine learning for rss-based visible light positioning ,” 2020. [Online]. Available: https://dx.doi.org/10.21227/f28n-6292

work page doi:10.21227/f28n-6292 2020
[17]

Image sensor-bas ed visible light positioning dataset,

C. Pereira, P . Fonseca, and L. Alves, “Image sensor-bas ed visible light positioning dataset,” 2023. [Online]. Avail able: https://dx.doi.org/10.21227/9zsx-bm77

work page doi:10.21227/9zsx-bm77 2023
[1]

A sur vey on indoor visible light positioning systems: Fundamentals , applications, and challenges,

Z. Zhu, Y . Y ang, M. Chen, C. Guo, J. Cheng, and S. Cui, “A sur vey on indoor visible light positioning systems: Fundamentals , applications, and challenges,” IEEE Communications Surveys & Tutorials , 2024. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2021 6

work page 2024
[2]

A m achine learning approach to ranging error mitigation for uwb local ization,

H. Wymeersch, S. Maran` o, W. M. Gifford, and M. Z. Win, “A m achine learning approach to ranging error mitigation for uwb local ization,” IEEE transactions on communications , vol. 60, no. 6, pp. 1719–1728, 2012

work page 2012
[3]

Genloc: A new paradigm for si gnal ﬁngerprinting with generative adversarial networks,

R. Guan, Y . Zhang, and M. Li, “Genloc: A new paradigm for si gnal ﬁngerprinting with generative adversarial networks,” in 2022 IEEE 12th International Conference on Indoor Positioning and Indoor Navigation (IPIN), 2022, pp. 1–8

work page 2022
[4]

Enh ancing rss- based visible light positioning by optimal calibration of l ed tilt and gain,

F. Wu, N. Stevens, L. D. Strycker, and F. Rottenberg, “Enh ancing rss- based visible light positioning by optimal calibration of l ed tilt and gain,” IEEE Transactions on Communications , pp. 1–1, 2025

work page 2025
[5]

Low- power centimeter-level localization for indoor mobile robots ba sed on ensemble kalman smoother using received signal strength,

Y . Zhuang, Q. Wang, M. Shi, P . Cao, L. Qi, and J. Y ang, “Low- power centimeter-level localization for indoor mobile robots ba sed on ensemble kalman smoother using received signal strength,” IEEE Internet of Things Journal, vol. 6, no. 4, pp. 6513–6522, 2019

work page 2019
[6]

An acoustic simulation framework to support indoor positioni ng and data driven signal processing assessments,

D. Delabie, C. Buyle, B. Cox, L. V an Der Perre, and L. De Str ycker, “An acoustic simulation framework to support indoor positioni ng and data driven signal processing assessments,” in 2023 31st European Signal Processing Conference (EUSIPCO) , 2023, pp. 261–265

work page 2023

Show all 25 references

[7]

An experimental testbed for the performance evaluation of optical time difference of arriv al based indoor positioning,

J. De Bruycker, J. Audenaert, S. R. Teli, S. Zvanovec, M. H . Conde, L. De Strycker, and N. Stevens, “An experimental testbed for the performance evaluation of optical time difference of arriv al based indoor positioning,” Journal of Lightwave Technology , vol. 42, no. 23, pp....

2024
[8]

Wideband diversit y in multipath channels with nonuniform power dispersion proﬁles,

M. Win, G. Chrisikos, and A. Molisch, “Wideband diversit y in multipath channels with nonuniform power dispersion proﬁles,” IEEE Transactions on Wireless Communications , vol. 5, no. 5, pp. 1014–1022, 2006

2006
[9]

Performance of rake reception in dense multipath channels: implications of spr eading band- width and selection diversity order,

M. Win, G. Chrisikos, and N. Sollenberger, “Performance of rake reception in dense multipath channels: implications of spr eading band- width and selection diversity order,” IEEE Journal on Selected Areas in Communications, vol. 18, no. 8, pp. 1516–1525, 2000

2000
[10]

A measurement-based statistical model for industrial ult ra-wideband channels,

J. Karedal, S. Wyne, P . Almers, F. Tufvesson, and A. F. Mo lisch, “A measurement-based statistical model for industrial ult ra-wideband channels,” IEEE Transactions on Wireless Communications, vol. 6, no. 8, pp. 3028–3037, 2007

2007
[11]

Nlos identi- ﬁcation and weighted least-squares localization for uwb sy stems using multipath channel statistics,

˙ I. G¨ uvenc ¸, C.-C. Chong, F. Watanabe, and H. Inamura, “Nlos identi- ﬁcation and weighted least-squares localization for uwb sy stems using multipath channel statistics,” EURASIP Journal on Advances in Signal Processing, vol. 2008, pp. 1–14, 2007

2008
[12]

Tightly coupled vlp/ins integrated navigation by inclination esti mation and blockage handling,

X. Sun, Y . Zhuang, X. Y ang, J. Huai, T. Huang, and D. Feng, “Tightly coupled vlp/ins integrated navigation by inclination esti mation and blockage handling,” Satellite Navigation , vol. 6, no. 1, p. 7, 2025

2025
[13]

Visible light positioning using a single led luminaire,

J. Hao, J. Chen, and R. Wang, “Visible light positioning using a single led luminaire,” IEEE Photonics Journal , vol. 11, no. 5, pp. 1–13, 2019

2019
[14]

A single led v isible light positioning system based on geometric features and cmos cam era,

H. Cheng, C. Xiao, Y . Ji, J. Ni, and T. Wang, “A single led v isible light positioning system based on geometric features and cmos cam era,” IEEE Photonics Technology Letters , vol. 32, no. 17, pp. 1097–1100, 2020

2020
[15]

Towards robust visible lig ht positioning under led shortage by visual-inertial fusion,

Q. Liang, J. Lin, and M. Liu, “Towards robust visible lig ht positioning under led shortage by visual-inertial fusion,” in 2019 International Conference on Indoor Positioning and Indoor Navigation (IP IN), 2019, pp. 1–8

2019
[18]

Data-efﬁcient gaussian process regression for accurate v isible light positioning,

N. Knudde, W. Raes, J. De Bruycker, T. Dhaene, and N. Stev ens, “Data-efﬁcient gaussian process regression for accurate v isible light positioning,” IEEE Communications Letters , vol. 24, no. 8, pp. 1705– 1709, 2020

2020
[19]

Data-efﬁcient ar tiﬁcial neural networks with gaussian process regression for 3d vis ible light positioning,

W. Zeng, H. Chen, J. Chen, and X. Hong, “Data-efﬁcient ar tiﬁcial neural networks with gaussian process regression for 3d vis ible light positioning,” in 2021 Optical Fiber Communications Conference and Exhibition (OFC) . IEEE, 2021, pp. 1–3

2021
[20]

A survey of positioning systems using visible l ed lights,

Y . Zhuang, L. Hua, L. Qi, J. Y ang, P . Cao, Y . Cao, Y . Wu, J. Thompson, and H. Haas, “A survey of positioning systems using visible l ed lights,” IEEE Communications Surveys & Tutorials , vol. 20, no. 3, pp. 1963– 1988, 2018

1963
[21]

Ackermann steering geometr y,

Wikipedia contributors, “Ackermann steering geometr y,” https://en.wikipedia.org/wiki/Ackermann steering geometry#cite note-1, 2025, [Online; accessed 23-Apr-2025]

2025
[22]

Luxeon IR Domed Line Infrared Emitters,

LumiLEDs, “Luxeon IR Domed Line Infrared Emitters,” https://lumileds.com/products/infrared-emitters/luxeon-ir-domed-line/ , 2025, [Online; accessed 23-Apr-2025]

2025
[23]

Visi ble light posi- tioning as a next-generation indoor positioning technolog y: A tutorial,

S. Bastiaens, M. Alijani, W. Joseph, and D. Plets, “Visi ble light posi- tioning as a next-generation indoor positioning technolog y: A tutorial,” IEEE Communications Surveys & Tutorials , vol. 26, no. 4, pp. 2867– 2913, 2024

2024
[24]

MIQUS Motion Capture Cameras,

Qualisys AB, “MIQUS Motion Capture Cameras,” https://www.qualisys.com/cameras/miqus/, 2025, [Online; accessed 23-Apr-2025]

2025
[25]

Inertial nav igation aiding by stationary updates,

A. Ramanandan, A. Chen, and J. A. Farrell, “Inertial nav igation aiding by stationary updates,” IEEE Transactions on Intelligent Transportation Systems, vol. 13, no. 1, pp. 235–248, 2011

2011

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [16]

Machine learning for rss-based visible light positioning ,

W. Raes, J. De Bruycker, N. Knudde, T. Dhaene, and N. Stev ens, “Machine learning for rss-based visible light positioning ,” 2020. [Online]. Available: https://dx.doi.org/10.21227/f28n-6292

work page doi:10.21227/f28n-6292 2020

[2] [17]

Image sensor-bas ed visible light positioning dataset,

C. Pereira, P . Fonseca, and L. Alves, “Image sensor-bas ed visible light positioning dataset,” 2023. [Online]. Avail able: https://dx.doi.org/10.21227/9zsx-bm77

work page doi:10.21227/9zsx-bm77 2023

[3] [1]

A sur vey on indoor visible light positioning systems: Fundamentals , applications, and challenges,

Z. Zhu, Y . Y ang, M. Chen, C. Guo, J. Cheng, and S. Cui, “A sur vey on indoor visible light positioning systems: Fundamentals , applications, and challenges,” IEEE Communications Surveys & Tutorials , 2024. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2021 6

work page 2024

[4] [2]

A m achine learning approach to ranging error mitigation for uwb local ization,

H. Wymeersch, S. Maran` o, W. M. Gifford, and M. Z. Win, “A m achine learning approach to ranging error mitigation for uwb local ization,” IEEE transactions on communications , vol. 60, no. 6, pp. 1719–1728, 2012

work page 2012

[5] [3]

Genloc: A new paradigm for si gnal ﬁngerprinting with generative adversarial networks,

R. Guan, Y . Zhang, and M. Li, “Genloc: A new paradigm for si gnal ﬁngerprinting with generative adversarial networks,” in 2022 IEEE 12th International Conference on Indoor Positioning and Indoor Navigation (IPIN), 2022, pp. 1–8

work page 2022

[6] [4]

Enh ancing rss- based visible light positioning by optimal calibration of l ed tilt and gain,

F. Wu, N. Stevens, L. D. Strycker, and F. Rottenberg, “Enh ancing rss- based visible light positioning by optimal calibration of l ed tilt and gain,” IEEE Transactions on Communications , pp. 1–1, 2025

work page 2025

[7] [5]

Low- power centimeter-level localization for indoor mobile robots ba sed on ensemble kalman smoother using received signal strength,

Y . Zhuang, Q. Wang, M. Shi, P . Cao, L. Qi, and J. Y ang, “Low- power centimeter-level localization for indoor mobile robots ba sed on ensemble kalman smoother using received signal strength,” IEEE Internet of Things Journal, vol. 6, no. 4, pp. 6513–6522, 2019

work page 2019

[8] [6]

An acoustic simulation framework to support indoor positioni ng and data driven signal processing assessments,

D. Delabie, C. Buyle, B. Cox, L. V an Der Perre, and L. De Str ycker, “An acoustic simulation framework to support indoor positioni ng and data driven signal processing assessments,” in 2023 31st European Signal Processing Conference (EUSIPCO) , 2023, pp. 261–265

work page 2023