DoRF++: Spherical Representation Learning over Doppler Radiance Fields for Robust Wi-Fi Sensing

Navid Hasanzadeh; Shahrokh Valaee

REVIEW 3 major objections 5 minor 27 references

DoRF++: Spherical Representation Learning over Doppler Radiance Fields for Robust Wi-Fi Sensing

T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Wi-Fi hand gestures can be recognized across unseen users from a single access point by reconstructing Doppler projections as views of a latent 3-D motion.

desk verdict DoRF++ is a well-ablated empirical paper with a plausible central claim, but the unvalidated linear-projection assumption and single-environment evidence keep it conditional. read the letter →

arxiv 2608.08381 v1 pith:PXS3ZQOZ submitted 2026-08-09 cs.CV eess.SP

Navid Hasanzadeh , Shahrokh Valaee This is my paper · ORCID

classification cs.CVeess.SP

keywords Wi-FisensingchannelstateinformationDopplerradiancefieldshumanactivityrecognitionsphericalTransformercross-usergeneralizationMUSICestimationhandgesture

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

The paper claims that the many Doppler velocity signals extracted from Wi-Fi CSI are not independent noisy descriptors but sparse, direction-specific observations of one underlying three-dimensional hand motion. It introduces DoRF, which recovers that latent motion by a regularized rank-three matrix factorization, then re-projects it onto a fixed sphere; DoRF++ adds spherical attention for classification. On a collected four-gesture dataset, this raises leave-one-subject-out accuracy from 66.4% to 80.4% for the best single receiver, and from 54.9% to 67.1% averaged over three receiver placements. If correct, the practical upshot is that privacy-preserving gesture recognition needs no wearable and no dense sensor array: a single multi-antenna router can do it.

What carries the argument

The load-bearing object is the Doppler Radiance Field tensor $\mathcal{P}\in\mathbb{R}^{T\times M\times 2M}$, formed by re-projecting the recovered latent velocity sequence $V\in\mathbb{R}^{T\times 3}$ onto an equiangular latitude-longitude grid of unit directions $d_{mn}$ via $P(s,m,n)=v(s)^\top d_{mn}$. The recovery step treats the observed Doppler projections $V_r$ as the product $VR+E$ of latent motion and unknown effective projection directions, solved by alternating ridge-regression updates; a global orthogonal ambiguity remains, which motivates the spherical learning stage. DoRF++ then treats each spherical direction as a token, extracts temporal features with fixed random convolution kernels, and applies spherical multi-head attention whose scores carry a relative positional bias built from Legendre polynomials of the cosine between directions and are weighted by spherical quadrature weights.

What would settle it

Take synthetic CSI generated from a known 3-D hand velocity with known projection directions, run MUSIC and DoRF, and check whether the recovered $V$ matches the ground-truth velocity up to a global orthogonal transform; in real data, record the same gestures with an optical motion-capture system alongside the Wi-Fi receiver and test whether dominant Doppler peaks and the recovered latent trajectory track the measured hand velocity. If classification accuracy stays high while the recovered motion fails to align with ground truth, the geometric mechanism is not the actual source of the reported gains.

Watch

Extended reading notes

Core claim

On its own terms, the paper claims that the seemingly chaotic multipath Doppler information inside Wi-Fi CSI can be organized into a coherent picture of the underlying body motion. For each receive antenna it forms ratios between transmit streams that share that antenna, removing receiver-side synchronization errors; from each ratio stream it extracts one dominant Doppler velocity per time window with MUSIC; then it solves a regularized rank-three factorization $V_r \approx VR$ for a latent 3-D velocity sequence $V$ and unknown effective projection directions $R$. The recovered motion is re-projected onto an equiangular sphere, and a spherical Transformer with quadrature-aware attention classifies the resulting Doppler radiance field. In the paper's leave-one-subject-out experiment, this raises single-receiver accuracy to 80.4% from 66.4% for the strongest prior method, with the biggest gains on gesture pairs that look alike from some viewpoints.

Load-bearing premise

The whole reconstruction rests on assuming that the dominant Doppler peak extracted from each common-RX, different-TX CSI-ratio stream is a linear projection of the same hand velocity onto a direction that stays fixed during the trial; if the peak follows some other scatterer or the effective direction rotates mid-trial, the recovered 3-D motion is a fitted artifact rather than true hand motion.

Editorial extensions

If this is right

A single multi-antenna Wi-Fi receiver is enough for practical cross-user gesture recognition, with accuracy tied to how directly the user lies on the dominant propagation path.
Unordered aggregation of Doppler projections leaves accuracy on the table; recovering a shared 3-D motion before classification is what unlocks the reported gain.
Because the global orientation of the recovered latent motion is arbitrary per trial, classifiers must be rotation-aware, and spherical attention plus max-pooling over antennas is the paper's proposed cure.
The largest remaining confusions are between gestures that become equivalent under a rotation of the coordinate axes, such as left-right versus push-pull, which is an intrinsic limit of Doppler-only sensing.
The ablation results indicate that the common-RX phase-sanitization step is responsible for the single largest accuracy swing, so its quality directly determines whether the rest of the pipeline has clean phase dynamics to work with.

Reading between the lines

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

Nothing in the method constrains the hand to be the only moving object; if the dominant MUSIC peak occasionally tracks a torso or a stationary reflector, the recovered latent motion would be a fitted artifact. Comparing recovered $V$ against optical motion capture in a controlled setup would make the geometric claim directly measurable; the paper stops at classification accuracy.
The same common-RX phase-sanitization and DoRF construction should transfer to other Doppler-capable RF sensors and to swapped transmitter-receiver roles, since the model only assumes linear projections of a latent 3-D velocity; this transfer is not tested here.
The dominant-peak truncation discards secondary Doppler peaks, which likely carry cross-user information for the confusing gesture pairs; a full-spectrum or multi-peak generalization of the factorization is a natural extension.
The per-trial orientation ambiguity means the recovered latent motion cannot yet be compared across trials without alignment; canonical orientation alignment would turn DoRF from a classifier input into a motion-capture-like representation for pose tracking.

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

Quick take: DoRF++ is a solid, honest empirical paper that makes a plausible case for treating Wi-Fi Doppler projections as low-rank views of a latent 3D velocity, then classifying a spherical re-projection with a spherical Transformer. The ablation is the real contribution: it isolates the representation from the classifier in a clean leave-one-subject-out protocol, and the full model beats the ablated variants by margins that look consistent. I came away convinced the pipeline does something useful on their dataset.

What's new: the common-RX CSI-ratio phase sanitization is a simple, sensible fix to a known problem; factoring Doppler projections into V and R is a reasonable extension of MORIC; and applying quadrature-aware spherical attention to the re-projected field is a natural fit. The relative positional bias sensibly addresses the orthogonal ambiguity. The paper is also candid: it explicitly says Proposition 2 is an ideal reference, that the recovered frame has an orthogonal ambiguity, and that spherical attention is only approximately equivariant. That's the right tone.

The soft spots are proportionate. The load-bearing assumption—dominant MUSIC peak per CSI-ratio stream is a linear projection of hand velocity onto a fixed effective direction—is deferred to MORIC and never validated against ground truth. If that fails, the recovered V is a fitted artifact. But the classification results don't depend on physical interpretability; the ablation shows the re-projection helps even if the latent motion isn't the true hand motion. More consequential is the evidence base: one room, one body orientation, ten participants, no significance tests, no code or data. AP3's gain is real but small and noisy. Also, the main comparison is to the authors' own MORIC; that's the right baseline, but independent replication would help. Self-citation to MORIC for the projection model is fine, yet the deferred derivation makes the core model hard to evaluate internally.

Who is this for? Wi-Fi sensing folks working on cross-user generalization, and anyone interested in spherical attention for structured representations. It deserves a serious referee—I'd send it out, not desk reject—but I'd condition acceptance on releasing data/code, adding significance testing, and ideally a second environment or a ground-truth check of the projection model.

Referee Report

3 major / 5 minor

Summary. This paper proposes DoRF++ for Wi-Fi-based human activity recognition. The pipeline first forms common-receive-antenna CSI ratios between transmit streams to suppress synchronization phase errors, estimates a single dominant Doppler velocity per ratio stream via MUSIC, and then fits, per receive antenna, a regularized rank-3 matrix factorization Vr ≈ VR that is interpreted as latent 3D hand velocity observed through unknown effective view directions. The recovered velocity sequence is re-projected onto a fixed equiangular spherical grid (the Doppler radiance field), and classification is performed by a spherical Transformer with quadrature-aware attention and rotation-invariant relative position biases. Experiments on a self-collected 10-user, 4-gesture dataset (UTHAMO-5G) with three receiver APs, under a leave-one-subject-out protocol, report that DoRF++ reaches 80.4% accuracy on AP1 versus 66.4% for MORIC and 67.1% averaged over APs versus 54.9% for MORIC; ablations (Table IV) attribute the gains to the common-RX phase sanitization, the DoRF factorization, and spherical attention.

Significance. The claimed gains are practically relevant if they hold: fine-grained cross-user hand-gesture recognition from a single Wi-Fi receiver AP is a hard and useful setting. The evaluation has several strengths: the LOSO protocol keeps training, validation, and test subjects disjoint; the ablation that feeds the DoRF representation into the MORIC classifier (DoRF vs MORIC in Table I) isolates the representation from the classifier; and the paper is unusually honest about the ideal nature of its identifiability result (Proposition 2) and the residual orthogonal ambiguity. The alternating-minimization convergence argument is standard but correctly stated. The main weaknesses are statistical: no significance tests are reported for the headline comparisons, and the 'state-of-the-art' claim rests on a narrow baseline set. The physical interpretation of the recovered latent velocity is not validated against ground truth, so the radiance-field framing currently outruns the evidence, although this does not invalidate the empirical classification claim.

major comments (3)

[§IV-C1, Table I] The central claim that DoRF++ 'significantly outperforms' prior methods is not backed by any statistical test. The ten LOSO folds are paired by test subject, so the authors should report paired tests (e.g., Wilcoxon signed-rank or paired t-test) for DoRF++ versus MORIC and DoRF++ versus DoRF, for each AP and for the AP-average, together with 95% confidence intervals. As reported, the AP1 gap (80.4±8.4 vs 66.4±7.8) and the AP3 gap (54.2±8.6 vs 42.4±6.8) overlap by roughly one standard deviation, so the reader cannot tell whether the gains are consistent across subjects or driven by a few folds. This is necessary to substantiate 'significantly outperforms' in the abstract and conclusion.
[§III-D, Eq. (23)-(27); Table IV] The latent motion V is never validated against ground-truth motion, yet the abstract states that DoRF 'infers a latent 3D motion sequence.' Because the spherical DoRF field P is a deterministic re-projection of V, the ablation results show that the rank-3 factorization and spherical re-projection help classification but do not establish physical interpretability. The projection model in Eq. (23) is asserted with the derivation deferred to MORIC [14], and Proposition 2 is explicitly ideal. I recommend either (i) validating V with motion capture or with synthetic Doppler data generated from known 3D trajectories, or (ii) rephrasing the contribution as a learned low-rank spherical representation inspired by radiance fields, rather than as recovery of the physical 3D motion. The current wording overstates the evidence.
[§IV-A, Table I] The 'state-of-the-art' comparison is too narrow to support the claim. The only recent strong baseline is MORIC [14], which is from the same group; Widar3.0 [13], cited in the introduction as a cross-domain gesture recognition method, is not included, nor are other independently developed 2023-2025 Wi-Fi HAR systems. The authors should add at least one independent recent baseline or soften the claim to 'outperforms the evaluated methods on UTHAMO-5G.'

minor comments (5)

[§III-A, Eq. (5)-(6)] The common-RX ratio cancels receiver-side STO/SFO but does not cancel TX-specific cyclic-shift-diversity delays or per-stream beamforming phases; the text should state that these residuals are static (hence Doppler-free) to avoid an imprecise claim about canceling all common phase terms.
[§III-D, Proposition 2] With four transmit antennas there are exactly six pairwise difference directions per receive antenna, and these six differences satisfy linear dependencies; the paper should comment on whether the spanning condition of Proposition 2 can actually be met under this construction, rather than only noting that N<6 is insufficient.
[§IV-B] Please state explicitly whether the DoRF matrix factorization is fit independently per trial (i.e., without using labels or other trials), since this is important for ruling out information leakage in the LOSO protocol.
[§IV-A] The 'mean±std' in Tables I and III should specify that the standard deviation is across the ten LOSO test subjects, and it would help to report per-fold results in a supplementary table.
[Eq. (5) and Table III] Minor presentation issues: the SFO term notation f'_c/f_c-1 in Eq. (5) is hard to parse, and the lower-triangular entries of the binary-accuracy matrix in Table III appear to be missing or misaligned in the typeset version.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: DoRF++'s accuracy gains come from supervised held-out-subject evaluation, and the self-cited MORIC projection model is a stated assumption rather than a fitted target.

full rationale

The paper's central claim is an empirical one: DoRF++ improves cross-user generalization accuracy on a hand-gesture dataset. This claim is evaluated with leave-one-subject-out cross-validation, with training, validation, and test subjects explicitly disjoint, and with external baselines (AMAP, CMAP, CapsHAR, CSI-ratio model, APNSS+APSC) in addition to MORIC. The DoRF representation itself is obtained by a regularized rank-three factorization of the observed Doppler projection matrix (Eqs. 26-27), and the spherical DoRF is explicitly described as a deterministic re-projection: the paper states, "This re-projection does not introduce new motion information beyond the recovered latent vector." Thus the spherical representation is a fitted feature extractor, not a redefinition of the accuracy metric, and the classifier is trained and tested on labels from held-out subjects. The main self-citation is MORIC [14], which supplies the linear-projection model used to interpret the Doppler observations (Eqs. 12-14 and 23). While this is load-bearing for the physical interpretation of the latent velocity V, the empirical classification result does not depend on V being physically meaningful hand velocity: the ablations show that the full DoRF++ construction outperforms direct latent-V, PCA-based, LSTM, and CNN alternatives, which is evidence that the advantage is not merely inherited from the fitted factorization. The paper also honestly disclaims identifiability: Proposition 2 is called "an ideal geometric reference" and the orthogonal ambiguity of the recovered factors is acknowledged. These are limitations in interpretability and identifiability, not circular derivation steps. Overall, no prediction in the paper reduces by construction to its inputs; at most there is a minor, non-load-bearing self-citation for the projection model.

Assumptions & free parameters 5 free parameters · 3 assumptions · 2 invented entities

The central model introduces two latent quantities, a 3D velocity time series V and stream-specific projection vectors R, both fitted to the Doppler observations they are supposed to explain. The key physical assumption, that a CSI-ratio stream's dominant Doppler bin is a linear projection of hand velocity, is deferred to a self-cited prior paper and never tested against ground truth. The remaining free parameters are hyperparameters of the fitting and the attention model.

free parameters (5)

velocity regularization mu = 0.1
Chosen by hand to stabilize the rank-3 factorization; controls the energy of the recovered latent velocity sequence and thus the shape of the DoRF.
direction regularization gamma = 0.01
Chosen by hand to prevent unstable effective projection vector estimates in the alternating minimization.
number of dominant Doppler components per stream = 1
MUSIC retains only the dominant peak per window, discarding all secondary Doppler components; this choice defines the scalar observations the DoRF model must explain.
Doppler search range and grid = -32 to 32 Hz, 513 points
Limits measurable velocities and sets frequency resolution; values chosen by hand for the 5 GHz setup.
equiangular spherical grid resolution M = 6
Determines the DoRF size (72 directions); chosen trade-off between resolution and token count for attention.

assumptions (3)

domain assumption Linear projection model: dominant MUSIC Doppler peak of a common-RX CSI-ratio stream equals v(s)^T (r_{m1,n} - r_{m2,n}) plus noise (Eq. 12-14).
The entire DoRF construction rests on this asserted relation. The paper defers the detailed derivation to self-cited MORIC [14] and provides no ground-truth validation, so if the dominant peak tracks a non-hand component or the ratio is not a single Doppler component, the factorization is fitting a spurious structure.
domain assumption Effective observation directions are constant over the whole trial: a single R is used for all time steps in Eq. (23).
The multipath geometry changes as the hand moves and as the body shifts, so the difference vector r_{m1,n} - r_{m2,n} may drift within a trial. The paper only states a local constancy approximation for short windows, then uses one global R for the entire trial.
domain assumption CSI impairment model in Eq. (5) with common and chain-specific STO/SFO decomposition.
The common-RX ratio phase sanitization argument assumes the cancellation works as written, that no subcarrier-dependent residual phase remains, and that the model captures the dominant impairments. This is standard in the field but is not verified against hardware measurements in this paper.

invented entities (2)

Latent 3D velocity sequence V
purpose: Shared motion representation whose linear projections explain all observed CSI-ratio Doppler velocities; the source of the Doppler radiance field.
V is recovered purely from the CSI data being explained. It is never compared with ground-truth hand motion or any external measurement, so it has no falsifiable handle outside the paper; its only support is that downstream classification improves.
Effective Doppler projection vectors r_i
purpose: Absorb the unknown, environment-dependent observation direction and gain of each CSI-ratio stream.
These vectors are inferred along with V from the same Doppler projections. Their normalized versions are used for spherical interpretation, but no external validation (e.g., measured angle-of-arrival or ray tracing) is provided.

how reviews work

0 comments

Cite this review

Pith. "Pith review of DoRF++: Spherical Representation Learning over Doppler Radiance Fields for Robust Wi-Fi Sensing." pith.science (2026). https://pith.science/paper/PXS3ZQOZ

@misc{pith2026260808381,
  author       = {Pith},
  title        = {Pith review of: DoRF++: Spherical Representation Learning over Doppler Radiance Fields for Robust Wi-Fi Sensing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PXS3ZQOZ}},
  note         = {Machine review of arXiv:2608.08381}
}

read the original abstract

Motivated by the IEEE 802.11bf effort to standardize advanced WLAN sensing, interest in Wi-Fi Channel State Information (CSI) for passive, device-free, and privacy-preserving activity and gesture recognition has grown rapidly. Recent studies have shown that Doppler velocity projections extracted from CSI, which directly reflect human-motion velocity, enable more robust human activity recognition (HAR) and stronger generalization across users and unseen conditions. Nevertheless, reliable generalization under real-world variability remains a major challenge, hindering the adoption of Wi-Fi sensing in real-world applications. To address this challenge, we introduce Doppler Radiance Fields (DoRF), bringing the concept of neural radiance fields (NeRF) from computer vision into Wi-Fi sensing. DoRF models Doppler velocity projections extracted from Wi-Fi CSI as sparse and diverse virtual-camera views of human motion. It then infers a latent 3D motion sequence whose projections along learned effective Doppler directions explain the CSI-derived Doppler observations. The recovered motion is subsequently projected onto an equiangular grid of directions on the unit sphere, producing a spherical representation of the underlying motion. Since DoRF naturally defines the Doppler representation on spheres, we further introduce DoRF++, a spherical-learning design that applies spherical Transformers for activity classification. Experiments on our collected hand-gesture dataset show that DoRF++ significantly outperforms state-of-the-art Wi-Fi-based HAR methods in cross-user generalization accuracy, especially for difficult gestures in settings with a single multi-antenna receiver access point (AP).

Figures

Figures reproduced from arXiv: 2608.08381 by the authors.

**Figure 2.** Proposed CSI phase sanitization based on the ratio of common-RX, [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Illustration of a DoRF constructed from Wi-Fi CSI. Sphere dots indicate [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗

Figures from the paper (4 more)

**Figure 5.** Figure 5 [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]

**Figure 4.** Figure 4: The proposed DoRF++ framework for Wi-Fi-based HAR. CSI signals are first transformed into Doppler velocity projections that capture motion through [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]

**Figure 5.** Figure 5: Illustration of the four right-hand gestures included in the UTHAMO [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]

**Figure 7.** Figure 7: Left–right and push–pull gestures can become equivalent under a [PITH_FULL_IMAGE:figures/full_fig_p014_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 21 canonical work pages

[14]

MORIC: CSI delay-doppler decomposi- tion for robust Wi-Fi-based human activity recognition,

N. Hasanzadeh and S. Valaee, “MORIC: CSI delay-doppler decomposi- tion for robust Wi-Fi-based human activity recognition,”arXiv preprint arXiv:2506.12997, 2025

arXiv 2025
[13]

Widar3.0: Zero-effort cross-domain gesture recognition with Wi-Fi,

Y . Zhang, Y . Zheng, K. Qian, G. Zhang, Y . Liu, C. Wu, and Z. Yang, “Widar3.0: Zero-effort cross-domain gesture recognition with Wi-Fi,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 11, pp. 8671–8688, 2021

work page 2021
[1]

Wi-Fi-based human sensing with deep learning: Recent advances, challenges, and opportunities,

I. Ahmad, A. Ullah, and W. Choi, “Wi-Fi-based human sensing with deep learning: Recent advances, challenges, and opportunities,”IEEE Open Journal of the Communications Society, 2024

work page 2024
[2]

An overview on IEEE 802.11bf: WLAN sensing,

R. Du, H. Hua, H. Xie, X. Song, Z. Lyu, M. Hu, Y . Xin, S. McCann, M. Montemurro, T. X. Hanet al., “An overview on IEEE 802.11bf: WLAN sensing,”IEEE Communications Surveys & Tutorials, 2024

work page 2024
[3]

A tutorial-cum-survey on self-supervised learning for Wi-Fi sensing: Trends, challenges, and outlook,

A. Y . Radwan, M. Yildirim, N. Hasanzadeh, H. Tabassum, and S. Valaee, “A tutorial-cum-survey on self-supervised learning for Wi-Fi sensing: Trends, challenges, and outlook,”IEEE Communications Surveys & Tutorials, 2025

work page 2025
[4]

Melon: NeRF with unposed images in SO(3),

A. Levy, M. Matthews, M. Sela, G. Wetzstein, and D. Lagun, “Melon: NeRF with unposed images in SO(3),”arXiv preprint arXiv:2303.08096, 2023

arXiv 2023
[5]

Beyond amplitude: Channel state information phase-aware deep fusion for robotic activity recogni- tion,

R. Zandi, H. Salehinejad, and M. Siami, “Beyond amplitude: Channel state information phase-aware deep fusion for robotic activity recogni- tion,”arXiv preprint arXiv:2603.09047, 2026

work page arXiv 2026
[6]

A survey on behavior recognition using Wi-Fi channel state information,

S. Yousefi, H. Narui, S. Dayal, S. Ermon, and S. Valaee, “A survey on behavior recognition using Wi-Fi channel state information,”IEEE Communications Magazine, vol. 55, no. 10, pp. 98–104, 2017

work page 2017

Show all 27 references

[7]

Fresnel zone- based voting with capsule networks for human activity recognition from channel state information,

R. Djogo, H. Salehinejad, N. Hasanzadeh, and S. Valaee, “Fresnel zone- based voting with capsule networks for human activity recognition from channel state information,”IEEE Internet of Things Journal, pp. 1–1, 2024. 17

2024
[8]

A study on the lightweight and fast response GRU techniques for indoor continuous motion recognition based on Wi-Fi CSI,

K. Jang, C. Sun, J. Zhou, Y . Seo, Y . Kim, and S. Choi, “A study on the lightweight and fast response GRU techniques for indoor continuous motion recognition based on Wi-Fi CSI,”IEEE Access, 2025

2025
[9]

CSI-GLSTN: A location-independent CSI human activity recognition method based on spatio-temporal and channel feature fusion,

Y . Zhang, D. Wu, Y . Wang, Y . Zhang, G. Ji, and J. Ai, “CSI-GLSTN: A location-independent CSI human activity recognition method based on spatio-temporal and channel feature fusion,”IEEE Transactions on Instrumentation and Measurement, 2024

2024
[10]

Rosefi: A robust sedentary behavior monitoring system with commodity Wi-Fi devices,

C. Peng, L. Gui, B. Sheng, Z. Guo, and F. Xiao, “Rosefi: A robust sedentary behavior monitoring system with commodity Wi-Fi devices,” IEEE Transactions on Mobile Computing, vol. 23, no. 5, pp. 6470–6489, 2023

2023
[11]

SHARP: Environment and person independent activity recognition with commodity IEEE 802.11 access points,

F. Meneghello, D. Garlisi, N. Dal Fabbro, I. Tinnirello, and M. Rossi, “SHARP: Environment and person independent activity recognition with commodity IEEE 802.11 access points,”IEEE Transactions on Mobile Computing, vol. 22, no. 10, pp. 6160–6175, 2022

2022
[12]

AFall: Wi- Fi-based device-free fall detection system using spatial angle of arrival,

S. Chen, W. Yang, Y . Xu, Y . Geng, B. Xin, and L. Huang, “AFall: Wi- Fi-based device-free fall detection system using spatial angle of arrival,” IEEE Transactions on Mobile Computing, vol. 22, no. 8, pp. 4471–4484, 2022

2022
[15]

DoRF: Doppler radiance fields for robust human activity recognition using Wi-Fi,

N. Hasanzadeh and S. Valaee, “DoRF: Doppler radiance fields for robust human activity recognition using Wi-Fi,” in2025 IEEE 10th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), 2025, pp. 156–160

2025
[16]

DoRF: Doppler radiance field-guided an- tenna selection for improved generalization in multi-antenna Wi-Fi-based human activity recognition,

N. Hasanzadeh and S. Valaee, “DoRF: Doppler radiance field-guided an- tenna selection for improved generalization in multi-antenna Wi-Fi-based human activity recognition,” inProc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2026

2026
[17]

NeRF: Representing scenes as neural radiance fields for view synthesis,

B. Mildenhall, P. P. Srinivasan, M. Tancik, J. T. Barron, R. Ramamoorthi, and R. Ng, “NeRF: Representing scenes as neural radiance fields for view synthesis,”Communications of the ACM, vol. 65, no. 1, pp. 99–106, 2021

2021
[18]

Wi-Fi sensing with channel state information: A survey,

Y . Ma, G. Zhou, and S. Wang, “Wi-Fi sensing with channel state information: A survey,”ACM Computing Surveys (CSUR), vol. 52, no. 3, pp. 1–36, 2019

2019
[19]

Wi-Fi CSI-based device-free sensing: From fresnel zone model to CSI-ratio model,

D. Wu, Y . Zeng, F. Zhang, and D. Zhang, “Wi-Fi CSI-based device-free sensing: From fresnel zone model to CSI-ratio model,”CCF Transactions on Pervasive Computing and Interaction, pp. 1–15, 2022

2022
[20]

Enhancing generalization in human activity recognition through improved Wi-Fi channel state information phase processing and antenna pair selection,

N. Hasanzadeh and S. Valaee, “Enhancing generalization in human activity recognition through improved Wi-Fi channel state information phase processing and antenna pair selection,” in2024 IEEE International Workshop on Machine Learning for Signal Processing (MLSP), 2024

2024
[21]

Multiple emitter location and signal parameter estimation,

R. Schmidt, “Multiple emitter location and signal parameter estimation,” IEEE Transactions on Antennas and Propagation, vol. 34, no. 3, pp. 276–280, 1986

1986
[22]

Doa estimation using a greedy block coordinate descent algorithm,

X. Wei, Y . Yuan, and Q. Ling, “Doa estimation using a greedy block coordinate descent algorithm,”IEEE Transactions on Signal Processing, vol. 60, no. 12, pp. 6382–6394, 2012

2012
[23]

Rocket: exceptionally fast and accurate time series classification using random convolutional kernels,

A. Dempster, F. Petitjean, and G. I. Webb, “Rocket: exceptionally fast and accurate time series classification using random convolutional kernels,” Data Mining and Knowledge Discovery, vol. 34, no. 5, pp. 1454–1495, 2020

2020
[24]

Joint human orientation-activity recognition using Wi-Fi signals for human-machine interaction,

H. Salehinejad, N. Hasanzadeh, R. Djogo, and S. Valaee, “Joint human orientation-activity recognition using Wi-Fi signals for human-machine interaction,” inICASSP 2023-2023 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE, 2023, pp. 1–5

2023
[25]

Attention on the sphere,

B. Bonev, M. Rietmann, A. Paris, A. Carpentieri, and T. Kurth, “Attention on the sphere,”arXiv preprint arXiv:2505.11157, 2025

2025 arXiv
[26]

Adam: A method for stochastic optimization,

D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” arXiv preprint arXiv:1412.6980, 2014. Navid Hasanzadehis currently a Ph.D. candidate in the Edward S. Rogers Sr. Department of Electri- cal and Computer Engineering at the University of Toronto, Toronto, ON,...

2014 arXiv
[2020]

He is currently a member of the Steering Committee of IEEE PIMRC

He was the co-chair of the organizing committee for PIMRC 2023. He is currently a member of the Steering Committee of IEEE PIMRC. From December 2010 to December 2012, he was the Associate Editor of the IEEE Signal Processing Letters. From 2010 to 2015, he served as an Editor o...

2023

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [14]

MORIC: CSI delay-doppler decomposi- tion for robust Wi-Fi-based human activity recognition,

N. Hasanzadeh and S. Valaee, “MORIC: CSI delay-doppler decomposi- tion for robust Wi-Fi-based human activity recognition,”arXiv preprint arXiv:2506.12997, 2025

arXiv 2025

[2] [13]

Widar3.0: Zero-effort cross-domain gesture recognition with Wi-Fi,

Y . Zhang, Y . Zheng, K. Qian, G. Zhang, Y . Liu, C. Wu, and Z. Yang, “Widar3.0: Zero-effort cross-domain gesture recognition with Wi-Fi,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 11, pp. 8671–8688, 2021

work page 2021

[3] [1]

Wi-Fi-based human sensing with deep learning: Recent advances, challenges, and opportunities,

I. Ahmad, A. Ullah, and W. Choi, “Wi-Fi-based human sensing with deep learning: Recent advances, challenges, and opportunities,”IEEE Open Journal of the Communications Society, 2024

work page 2024

[4] [2]

An overview on IEEE 802.11bf: WLAN sensing,

R. Du, H. Hua, H. Xie, X. Song, Z. Lyu, M. Hu, Y . Xin, S. McCann, M. Montemurro, T. X. Hanet al., “An overview on IEEE 802.11bf: WLAN sensing,”IEEE Communications Surveys & Tutorials, 2024

work page 2024

[5] [3]

A tutorial-cum-survey on self-supervised learning for Wi-Fi sensing: Trends, challenges, and outlook,

A. Y . Radwan, M. Yildirim, N. Hasanzadeh, H. Tabassum, and S. Valaee, “A tutorial-cum-survey on self-supervised learning for Wi-Fi sensing: Trends, challenges, and outlook,”IEEE Communications Surveys & Tutorials, 2025

work page 2025

[6] [4]

Melon: NeRF with unposed images in SO(3),

A. Levy, M. Matthews, M. Sela, G. Wetzstein, and D. Lagun, “Melon: NeRF with unposed images in SO(3),”arXiv preprint arXiv:2303.08096, 2023

arXiv 2023

[7] [5]

Beyond amplitude: Channel state information phase-aware deep fusion for robotic activity recogni- tion,

R. Zandi, H. Salehinejad, and M. Siami, “Beyond amplitude: Channel state information phase-aware deep fusion for robotic activity recogni- tion,”arXiv preprint arXiv:2603.09047, 2026

work page arXiv 2026

[8] [6]

A survey on behavior recognition using Wi-Fi channel state information,

S. Yousefi, H. Narui, S. Dayal, S. Ermon, and S. Valaee, “A survey on behavior recognition using Wi-Fi channel state information,”IEEE Communications Magazine, vol. 55, no. 10, pp. 98–104, 2017

work page 2017