LiveSense: A Real-Time Wi-Fi Sensing Platform for Range-Doppler on COTS Laptop

Cagri Tanriover; Jessica Sanson; Maximilian Pinaroc; Rahul C. Shah; Valerio Frascolla

arxiv: 2603.06545 · v2 · submitted 2026-03-06 · 📡 eess.SP · cs.AI

LiveSense: A Real-Time Wi-Fi Sensing Platform for Range-Doppler on COTS Laptop

Jessica Sanson , Rahul C. Shah , Maximilian Pinaroc , Cagri Tanriover , Valerio Frascolla This is my paper

Pith reviewed 2026-05-15 14:47 UTC · model grok-4.3

classification 📡 eess.SP cs.AI

keywords Wi-Fi sensingChannel State InformationRange-DopplerCOTS hardwareReal-time sensingMicro-motion detectionSelf-interference cancellation

0 comments

The pith

Unmodified laptop Wi-Fi cards can deliver centimeter-level range and Doppler sensing in real time while keeping normal communication active.

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

LiveSense shows how a standard laptop's Intel AX211 or BE201 Wi-Fi card can be turned into a sensor that measures target distances and velocities down to centimeters. The system pulls synchronized channel state information at 40 Hz or faster, aligns phases and cancels self-interference directly on the device, then streams range, Doppler, and video data to a GUI. This works even with the 160 MHz bandwidth limit of consumer Wi-Fi and without any hardware changes or external calibration. A reader would care because it demonstrates that everyday laptops could perform useful sensing tasks like detecting breathing or gestures without buying extra equipment.

Core claim

LiveSense is the first demonstration that accurate range information of targets can be obtained from commercial Wi-Fi despite the limited 160 MHz bandwidth, achieved by extracting fully-synchronized CSI at rates of 40 Hz and higher from COTS Intel AX211 or BE201 NICs, followed by on-device time-phase alignment and self-interference cancellation.

What carries the argument

On-device CSI extraction and processing pipeline that performs time-phase alignment and self-interference cancellation on unmodified Intel Wi-Fi 6E/7 NICs.

If this is right

Real-time detection of distance and radial velocity of people within a few meters.
Sensing of micro-motions such as respiration.
Ranging and tracking for hand gestures.
Simultaneous streaming of range, Doppler, subcarrier magnitude/phase, and annotated video to a Python/Qt GUI.
Continued normal Wi-Fi communication alongside sensing.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Existing consumer laptops could be repurposed for contactless monitoring of vital signs or activity in homes and offices without added hardware.
The approach could be tested in multi-laptop networks to see whether data fusion improves localization accuracy beyond single-device performance.
If the phase information proves stable across different environments, the method might extend to other Wi-Fi standards or frequency bands with minimal retuning.

Load-bearing premise

Channel state information extracted from unmodified commercial Wi-Fi cards contains enough phase and timing detail to support centimeter-level range estimation after only on-device processing and without external calibration.

What would settle it

A controlled measurement in which LiveSense range estimates for a static target at known distance differ by more than a few centimeters from simultaneous readings by a laser rangefinder or tape measure.

read the original abstract

We present LiveSense - a cross-platform that transforms a commercial off-the-shelf (COTS) Wi-Fi Network Interface Card (NIC) on a laptop into a centimeter-level Range-Doppler sensor while preserving simultaneous communication capability. The laptops are equipped with COTS Intel AX211 (Wi-Fi 6E) or Intel BE201 (Wi-Fi 7) NICs. LiveSense can (i) Extract fully-synchronized channel state information (CSI) at >= 40 Hz, (ii) Perform time-phase alignment and self-interference cancellation on-device, and (iii) Provide a real-time stream of range, Doppler, subcarrier magnitude/phase and annotated video frames to a Python/Qt Graphical User Interface (GUI). The demo will showcase the ability to detect (i) Distance and radial velocity of attendees within a few meters of the device, (ii) Micro-motion (respiration), and (iii) Hand-gesture ranging. To the best of our knowledge, this is the first-ever demo to obtain accurate range information of targets from commercial Wi-Fi, despite the limited 160 MHz bandwidth.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

LiveSense puts together a working real-time range-Doppler pipeline on unmodified Intel laptop Wi-Fi cards, but the centimeter accuracy claim needs quantitative phase and error data to hold up.

read the letter

This paper shows a platform that extracts CSI at 40 Hz or better from standard AX211 or BE201 cards, runs on-device alignment and self-interference cancellation, and streams range, Doppler, and video to a GUI while the link stays active for normal traffic. The demos cover nearby person ranging, respiration, and hand gestures, which is the practical piece that matters for application work. That combination on unmodified COTS hardware is the concrete step forward here. The implementation details for cross-platform extraction and real-time processing are the parts that could actually be reused by others. The soft spot is the accuracy side. The claim of accurate range despite the 160 MHz bandwidth limit depends on subcarrier phase staying stable enough for time-of-flight at the centimeter scale. Consumer NICs commonly carry residual CFO, SFO, and packet-to-packet phase jumps that on-device fixes may not fully remove. Without ground-truth error distributions, phase-noise statistics, or direct comparisons in the text, it is difficult to tell whether the system reaches the stated precision or mainly shows detectable motion. The stress-test concern about needing roughly 0.2 rad phase stability lands directly on the central result. Readers who build sensing applications or want a ready starting point for Wi-Fi-based health or gesture work will find the platform description useful. It is a systems and demo contribution rather than a new derivation. I would send it for peer review. The implementation is solid enough to deserve referee time, mainly to tighten the validation numbers.

Referee Report

2 major / 1 minor

Summary. The manuscript presents LiveSense, a real-time Wi-Fi sensing platform that converts unmodified COTS Intel AX211 (Wi-Fi 6E) or BE201 (Wi-Fi 7) NICs on laptops into centimeter-level range-Doppler sensors while preserving simultaneous communication. It extracts fully-synchronized CSI at rates of at least 40 Hz, performs on-device time-phase alignment and self-interference cancellation, and streams range, Doppler, subcarrier magnitude/phase, and annotated video to a Python/Qt GUI. The work demonstrates detection of target distance and radial velocity within a few meters, micro-motion such as respiration, and hand-gesture ranging, claiming this as the first demo to achieve accurate range information from commercial Wi-Fi despite the 160 MHz bandwidth limit.

Significance. If the claimed centimeter-level range accuracy holds on unmodified hardware, the result would be significant for enabling practical, real-time Wi-Fi sensing applications (e.g., vital signs and gesture recognition) on ubiquitous laptop hardware without external calibration or specialized equipment. The cross-platform, on-device implementation that maintains communication capability is a practical strength that could broaden adoption of Wi-Fi sensing.

major comments (2)

[Abstract] Abstract: The central claim of obtaining 'accurate range information' at 'centimeter-level' despite 160 MHz bandwidth is not supported by any quantitative validation, error metrics, ground-truth comparisons, or phase-stability statistics. Standard range resolution from 160 MHz is ~0.94 m, so cm-scale accuracy requires explicit evidence that subcarrier phase differences remain stable to ≲0.2 rad after on-device processing; this evidence is absent.
[Implementation] Implementation section (on-device processing): The description of time-phase alignment and self-interference cancellation does not detail how residual CFO, SFO, and per-packet phase jumps inherent to Intel AX211/BE201 CSI are removed to the precision needed for usable ToF estimates; without this, the assumption that unmodified NIC CSI suffices for cm-level ranging cannot be evaluated.

minor comments (1)

[Abstract] Abstract: The sentence 'We present LiveSense - a cross-platform that transforms' contains a grammatical issue; rephrase for clarity (e.g., 'a cross-platform platform' or 'a platform').

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive review. The comments correctly identify areas where the current manuscript lacks sufficient quantitative support and implementation detail. We will revise the paper to address both points while preserving its focus as a real-time demonstration platform.

read point-by-point responses

Referee: [Abstract] Abstract: The central claim of obtaining 'accurate range information' at 'centimeter-level' despite 160 MHz bandwidth is not supported by any quantitative validation, error metrics, ground-truth comparisons, or phase-stability statistics. Standard range resolution from 160 MHz is ~0.94 m, so cm-scale accuracy requires explicit evidence that subcarrier phase differences remain stable to ≲0.2 rad after on-device processing; this evidence is absent.

Authors: We agree that the abstract's claim of centimeter-level accuracy is not backed by quantitative metrics in the submitted version. As the work is presented as a live demonstration of a cross-platform sensing system rather than a full measurement study, the emphasis was placed on real-time operation and GUI streaming. We will add a new subsection with controlled experiments reporting range and velocity error statistics against ground truth (laser and motion-capture references), phase-stability histograms across subcarriers, and ToF precision after alignment. These additions will directly support the accuracy claim. revision: yes
Referee: [Implementation] Implementation section (on-device processing): The description of time-phase alignment and self-interference cancellation does not detail how residual CFO, SFO, and per-packet phase jumps inherent to Intel AX211/BE201 CSI are removed to the precision needed for usable ToF estimates; without this, the assumption that unmodified NIC CSI suffices for cm-level ranging cannot be evaluated.

Authors: The referee is correct that the current description of on-device processing is high-level. We will expand the Implementation section with a step-by-step account of the alignment pipeline, including: (1) the pilot-based CFO/SFO estimation and compensation applied per packet, (2) the reference-antenna phase-jump removal procedure, and (3) the self-interference cancellation filter coefficients. We will also include measured residual phase variance after each stage to allow readers to assess whether the precision is adequate for the reported ranging performance. revision: yes

Circularity Check

0 steps flagged

No significant circularity; implementation demo without derived predictions or self-referential reductions

full rationale

The paper presents LiveSense as a real-time platform and demo that extracts CSI from unmodified Intel AX211/BE201 NICs, performs on-device time-phase alignment and self-interference cancellation, and streams range-Doppler outputs. No equations, first-principles derivations, or predictions are shown that reduce to fitted parameters, self-citations, or ansatzes by construction. The central claim is an empirical demonstration of cm-level ranging despite 160 MHz bandwidth limits, supported by implementation details rather than any load-bearing mathematical chain. This is a standard self-contained demo paper; the reader's assessment of score 2.0 is consistent with the absence of any circular steps.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No free parameters, axioms, or invented entities are identifiable from the abstract alone; the work is presented as an engineering platform rather than a theoretical derivation.

pith-pipeline@v0.9.0 · 5517 in / 991 out tokens · 28101 ms · 2026-05-15T14:47:20.263261+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/AbsoluteFloorClosure.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

CSI Cross-Correlation for coarse frame alignment... Per-Frame Phase Unwrapping... Adaptive Self-Interference Cancellation... 2-D FFT / DFT... CFAR detector

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.