pith. sign in

arxiv: 2604.07032 · v3 · submitted 2026-04-08 · 📡 eess.SP

Reliable Non-Line-of-Sight Intrusion Detection with Integrated Sensing and Communications Hardware

Paolo Tosi , Maximilian Bauhofer , Marcus Henninger , Laurent Schmalen , Silvio Mandelli This is my paper

Pith reviewed 2026-05-10 17:28 UTC · model grok-4.3

classification 📡 eess.SP
keywords non-line-of-sightintrusion detectionintegrated sensing and communicationsmillimeter-wave sensingtarget trackingKalman filterindustrial applications
0
0 comments X

The pith

A millimeter-wave integrated sensing and communications system detects moving targets behind obstructions in industrial environments using surface reflections.

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

This paper shows how to use existing communications hardware for sensing in areas without direct visibility, which could help secure factories and other indoor spaces. The approach processes reflected millimeter-wave signals from a large surface and uses tracking algorithms to identify real target movements while ignoring clutter. A sympathetic reader would care because it adds new capabilities to future wireless networks without needing separate sensors. Tests confirm the method works reliably and resists false detections from noise-like signals.

Core claim

The system leverages reflections off a large surface to detect moving targets in cluttered indoor industrial scenarios where the direct line-of-sight is obstructed. A signal processing pipeline with a tracking stage compares Kalman Filter and probability hypothesis density filter approaches to detect targets and track movements in non-line-of-sight conditions. Experimental results demonstrate reliable detection of target presence while avoiding false alarms, with additional tests using synthetic false peaks confirming robustness.

What carries the argument

The reflection-based non-line-of-sight sensing pipeline that applies tracking filters to millimeter-wave signals to distinguish targets from clutter in obstructed environments.

Load-bearing premise

Reflections off a single large surface together with the tracking filters will work in any cluttered indoor industrial scenario regardless of target speeds or clutter levels.

What would settle it

Finding a setup with high clutter density where the system frequently misses targets or triggers false alarms would show the approach does not generalize as claimed.

Figures

Figures reproduced from arXiv: 2604.07032 by Laurent Schmalen, Marcus Henninger, Maximilian Bauhofer, Paolo Tosi, Silvio Mandelli.

Figure 1
Figure 1. Figure 1: Proposed signal processing pipeline. Operations performed by the fifth generation (5G)-compliant communication hardware are [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Measurement scenario in the ARENA2036. The distance between the monostatic sensing setup and the cargo gate (highlighted in pink) [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Example output of the TDD peak detection step, before [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: KF-based track confirmation system. Incoming peaks [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Top: PHD and KF range estimates (blue lines) for the full [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Detection performance (FN rate vs. FP rate) for [PITH_FULL_IMAGE:figures/full_fig_p005_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Detection performance for varying observation durations [PITH_FULL_IMAGE:figures/full_fig_p006_7.png] view at source ↗
read the original abstract

Non-line-of-sight (NLOS) sensing has the potential to enable use cases like intrusion detection in occluded areas, increasing the value provided by Integrated Sensing and Commu- nications (ISAC) in future 6G cellular networks. In this paper, we present a reliable NLOS intrusion detection system based on a millimeter-wave ISAC proof-of-concept. By leveraging reflections off a large surface, the proposed system addresses the challenge of detecting moving targets in cluttered indoor industrial scenarios where the direct line-of-sight is obstructed. A signal processing pipeline including a tracking stage, comparing a Kalman Filter (KF)- and a probability hypothesis density (PHD) filter-based approach, is applied to detect targets and track movements in NLOS. Experimental validation conducted in the ARENA2036 industrial research campus demonstrates that our system can reliably detect target presence in NLOS while avoiding false alarms. Tests with synthetically generated false peaks further demonstrate the robustness of our system to false alarms. Overall, the results underline the potential of NLOS ISAC as a promising technology for enabling intrusion detection and monitoring use cases.

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.

Referee Report

3 major / 2 minor

Summary. The paper proposes a millimeter-wave ISAC proof-of-concept system for reliable non-line-of-sight (NLOS) intrusion detection in cluttered indoor industrial environments. It exploits reflections from a single large surface to enable sensing of moving targets when direct LOS is obstructed, applies a signal processing pipeline with Kalman Filter (KF) and Probability Hypothesis Density (PHD) tracking stages, and reports experimental validation in the ARENA2036 campus demonstrating target detection while avoiding false alarms, including robustness tests with synthetically injected false peaks.

Significance. If the experimental claims are supported by quantitative metrics, the work would provide a concrete demonstration of NLOS ISAC for practical 6G use cases such as intrusion detection and monitoring in occluded industrial spaces. The side-by-side comparison of KF and PHD filters in a real hardware setup is a useful contribution, and the use of synthetic false-peak injection to probe robustness is a positive methodological step. However, the single-environment validation limits the assessed impact on broader deployment.

major comments (3)
  1. Abstract and §5 (Experimental Validation): the headline claim of 'reliable' NLOS detection with 'avoiding false alarms' is presented without any reported detection probability (Pd), false-alarm probability (Pfa), ROC curves, trial counts, or error bars. This absence prevents assessment of whether the results meet the reliability threshold asserted in the abstract.
  2. §5 and §4 (Signal Processing Pipeline): the robustness demonstration relies on synthetically generated false peaks, yet the manuscript provides no quantitative comparison of how these peaks were generated or how their statistics relate to measured multipath in the ARENA2036 geometry. Without this mapping, it is unclear whether the KF/PHD filters' reported performance would hold under real varying target speeds or higher clutter densities.
  3. §5: the experimental campaign is confined to reflections from one large surface in a single campus layout. The central reliability claim therefore rests on an untested assumption that this geometry and clutter density are representative; no additional reflector configurations or clutter-density sweeps are reported to support generalization to arbitrary industrial NLOS scenarios.
minor comments (2)
  1. Figure captions and axis labels in the experimental results should explicitly state the number of trials and any data-exclusion criteria used.
  2. Notation for the PHD filter parameters (e.g., birth intensity, survival probability) should be defined consistently with the KF covariance matrices for direct comparison.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive feedback on our proof-of-concept mm-wave ISAC NLOS intrusion detection system. We have revised the manuscript to strengthen the quantitative support for our claims and to clarify methodological details. Each major comment is addressed below.

read point-by-point responses

  1. Referee: Abstract and §5 (Experimental Validation): the headline claim of 'reliable' NLOS detection with 'avoiding false alarms' is presented without any reported detection probability (Pd), false-alarm probability (Pfa), ROC curves, trial counts, or error bars. This absence prevents assessment of whether the results meet the reliability threshold asserted in the abstract.

    Authors: We agree that explicit metrics are needed to substantiate the reliability claims. The revised manuscript now reports the number of independent experimental trials conducted in the ARENA2036 environment, provides estimated Pd and Pfa values derived from the observed detections and absence of false alarms across those trials, and includes ROC curves for both the KF and PHD trackers in Section 5, with error bars indicating variability. These additions directly address the assessment of the headline claims. revision: yes

  2. Referee: §5 and §4 (Signal Processing Pipeline): the robustness demonstration relies on synthetically generated false peaks, yet the manuscript provides no quantitative comparison of how these peaks were generated or how their statistics relate to measured multipath in the ARENA2036 geometry. Without this mapping, it is unclear whether the KF/PHD filters' reported performance would hold under real varying target speeds or higher clutter densities.

    Authors: We thank the referee for highlighting this gap. The synthetic peaks were generated to match the amplitude, range, and Doppler statistics of clutter peaks extracted from the real ARENA2036 measurements. The revised Sections 4 and 5 now include the exact generation parameters, a quantitative comparison (including histograms and statistical moments) between synthetic and measured multipath, and a brief analysis of how the filters respond to increased densities. This mapping supports the robustness claims under the observed conditions while noting that extreme variations would require further study. revision: yes

  3. Referee: §5: the experimental campaign is confined to reflections from one large surface in a single campus layout. The central reliability claim therefore rests on an untested assumption that this geometry and clutter density are representative; no additional reflector configurations or clutter-density sweeps are reported to support generalization to arbitrary industrial NLOS scenarios.

    Authors: We acknowledge that validation in a single geometry limits broad generalization claims. The ARENA2036 setup was selected as a representative cluttered industrial environment with a dominant large reflector. The revised manuscript adds an explicit discussion in Section 5 of the geometric assumptions, expected applicability to similar industrial layouts, and the proof-of-concept scope. Additional reflector configurations or density sweeps would require a separate experimental campaign beyond the current work. revision: partial

Circularity Check

0 steps flagged

No circularity: experimental validation paper with independent empirical results

full rationale

The paper presents an ISAC-based NLOS intrusion detection system using a signal processing pipeline (KF/PHD tracking) and validates it via experiments in the ARENA2036 campus, including synthetic false-peak injection. No derivation chain, first-principles predictions, or fitted parameters are claimed that reduce to inputs by construction. The central claim rests on direct empirical demonstration of detection performance rather than any self-referential modeling step, self-citation load-bearing theorem, or ansatz smuggled via prior work. This is the most common honest outcome for hardware/experimental papers.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The paper is an experimental system demonstration and introduces no new mathematical axioms, free parameters, or invented physical entities; it applies standard Kalman and PHD filters to ISAC reflections.

pith-pipeline@v0.9.0 · 5503 in / 1126 out tokens · 31492 ms · 2026-05-10T17:28:11.418431+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

14 extracted references · 14 canonical work pages

  1. [1]

    Survey on integrated sensing and communication performance modeling and use cases feasibility,

    S. Mandelli, M. Henninger, M. Bauhofer, and T. Wild, “Survey on integrated sensing and communication performance modeling and use cases feasibility,” inInt. Conf. on 6G Netw., Oct. 2023

    work page 2023

  2. [2]

    Feasibility study on integrated sensing and communication,

    3GPP, “Feasibility study on integrated sensing and communication,” Technical Report (TR) 22.837, 2024, version 19.4.0

    work page 2024

  3. [3]

    Integrated sensing and communication: Enabling techniques, applications, tools and data sets, standardization, and future directions,

    J. Wanget al., “Integrated sensing and communication: Enabling techniques, applications, tools and data sets, standardization, and future directions,”IEEE Internet of Things Journal, vol. 9, no. 23, 2022

    work page 2022

  4. [4]

    A unified future: Integrated sensing and communication (isac) in 6g,

    A. Ghoshet al., “A unified future: Integrated sensing and communication (isac) in 6g,”IEEE J. of Sel. Topics in Electromagn., Antennas and Propag., vol. 1, no. 1, 2025

    work page 2025

  5. [5]

    Radar scheme with raised reflector for nlos vehicle detection,

    D. Solomitckii, M. Heino, S. Buddappagari, M. A. Hein, and M. Valkama, “Radar scheme with raised reflector for nlos vehicle detection,”IEEE Trans. on Intell. Transp. Syst., vol. 23, no. 7, pp. 9037–9045, 2022

    work page 2022

  6. [6]

    Millimeter-wave automotive radar scheme with passive reflector for blind corner conditions,

    D. Solomitckii, C. B. Barneto, M. Turunen, M. All ´en, Y . Koucheryavy, and M. Valkama, “Millimeter-wave automotive radar scheme with passive reflector for blind corner conditions,” in2020 14th European Conference on Antennas and Propagation (EuCAP), 2020, pp. 1–5

    work page 2020

  7. [7]

    Extraction of human micro-doppler signature in an urban environment using a “sensing-behind-the-corner

    M. Gustafsson, A. Andersson, T. Johansson, S. Nilsson, A. Sume, and A. ¨Orbom, “Extraction of human micro-doppler signature in an urban environment using a “sensing-behind-the-corner” radar,”IEEE Geosci. and Remote Sens. Lett., vol. 13, no. 2, pp. 187–191, 2016

    work page 2016

  8. [8]

    Detection of human breathing in non-line-of-sight region by using mmwave fmcw radar,

    G. Li, Y . Ge, Y . Wang, Q. Chen, and G. Wang, “Detection of human breathing in non-line-of-sight region by using mmwave fmcw radar,” IEEE Trans. on Instrum. and Meas., vol. 71, pp. 1–11, 2022

    work page 2022

  9. [9]

    Feasibility of non-line-of-sight integrated sensing and communication at mmwave,

    P. Tosi, M. Henninger, L. G. de Oliveira, and S. Mandelli, “Feasibility of non-line-of-sight integrated sensing and communication at mmwave,” in 2024 IEEE 25th International Workshop on Signal Processing Advances in Wireless Communications (SPA WC), 2024, pp. 331–335

    work page 2024

  10. [10]

    6g integrated sensing and communication: From vision to realization,

    T. Wild, A. Grudnitsky, S. Mandelli, M. Henninger, J. Guan, and F. Schaich, “6g integrated sensing and communication: From vision to realization,” in2023 20th Eur . Radar Conf., Sep. 2023, pp. 355–358

    work page 2023

  11. [11]

    Nr; physical channels and modulation,

    “Nr; physical channels and modulation,” 3GPP, Technical Specification (TS) 38.211, 2025, version 19.1.0

    work page 2025

  12. [12]

    M. A. Richards, J. A. Scheer, and W. A. Holm,Principles of Modern Radar: Basic Principles. Raleigh, NC, USA: SciTech Pub., 2010

    work page 2010

  13. [13]

    Target detection for isac with tdd transmission,

    M. Henninger, L. Giroto, S. Saur, A. Grudnitsky, T. Wild, and S. Mandelli, “Target detection for isac with tdd transmission,” in2025 28th International Workshop on Smart Antennas (WSA), 2025, pp. 1–7

    work page 2025

  14. [14]

    The gaussian mixture probability hypothesis density filter,

    B.-N. V o and W.-K. Ma, “The gaussian mixture probability hypothesis density filter,”IEEE Trans. Signal Process., vol. 54, no. 11, pp. 4091–4104, 2006

    work page 2006