Auto-correlation Function Keying

Fan Liu; Jianhua Zhang; Shuangyang Li; Weijiang Zhao; Yifeng Xiong

Reviewed by Pith at T0; open to challenge.

T0 means a machine referee read the full paper against a public rubric. The mark states how deep the mechanical check went, never who wrote it. the ladder, T0–T4 →

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

T0 review · grok-4.3

Embedding data symbols on ACF sidelobes enables exact control of nominal periodic auto-correlation peaks in ISAC.

2026-06-26 22:37 UTC pith:VA7VC6SM

load-bearing objection ACFK keys data onto ACF sidelobes for nominal peak control in ISAC signals, but the exact-control claim depends on a non-negativity condition whose typical satisfaction rate is not shown. the 2 major comments →

arxiv 2606.17970 v2 pith:VA7VC6SM submitted 2026-06-16 cs.IT math.IT

Auto-correlation Function Keying

Weijiang Zhao , Yifeng Xiong , Fan Liu , Shuangyang Li , Jianhua Zhang This is my paper

classification cs.IT math.IT

keywords ACFKISACperiodic auto-correlation functionpeak sidelobe levelmutual informationmodulation design6G

verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper develops auto-correlation function keying to create signals that carry data while controlling the peak sidelobes of their periodic auto-correlation function for sensing. It maximizes mutual information subject to peak sidelobe constraints and shows that a uniform ACF-domain construction is optimal at high SNR for flat channels. ACFK implements this by placing symbols on the sidelobes, giving exact nominal control that matches the actual function under spectral non-negativity. When the constraint fails, it quantifies the violation probability and bounds the sidelobe degradation, leading to better performance than probabilistic amplitude shaping in simulations.

Core claim

ACFK embeds data symbols directly onto the ACF-domain sidelobes to achieve exact control of the nominal P-ACF, which coincides with the actual P-ACF when the spectral non-negativity constraint is satisfied; otherwise the non-negativity violation probability is quantified and PSLR degradation bounded.

What carries the argument

Auto-correlation function keying (ACFK), a modulation architecture that embeds data symbols directly onto the ACF-domain sidelobes to control the periodic auto-correlation function.

Load-bearing premise

The spectral non-negativity constraint holds so that nominal P-ACF control matches the actual one, or the optimality applies only to quasi-static frequency-flat channels at high SNR.

What would settle it

Generate many ACFK signal realizations, compute their actual P-ACF peak sidelobe levels, and check if the degradation exceeds the derived bound when the spectrum goes negative in some realizations.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

Stronger PSLR control than generalized PAS baseline
Improved weak-target detection performance under comparable settings
High-SNR approximate BER analysis for the ISAC transceiver over multipath channels
Exact nominal P-ACF control when spectral non-negativity holds

Where Pith is reading between the lines

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

The ACFK approach could be tested in frequency-selective channels to see if the control extends beyond flat fading assumptions.
Connecting ACFK to other correlation-based modulations might reveal broader applications in radar and communications.
The bounds on PSLR degradation could be tightened with more detailed spectral analysis.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

ACFK keys data onto ACF sidelobes for nominal peak control in ISAC signals, but the exact-control claim depends on a non-negativity condition whose typical satisfaction rate is not shown.

read the letter

The main takeaway is that ACFK gives a modulation scheme that puts data directly on the sidelobes of the periodic auto-correlation to control peak levels exactly when a spectral condition holds, and this looks like a genuine step beyond average-based methods in ISAC.

What stands out as new is the ACFK architecture itself, which embeds symbols onto ACF-domain sidelobes for nominal P-ACF control. The paper also derives an asymptotic optimality result for the continuous version under high-SNR flat channels and supplies bounds on how much PSLR worsens when the non-negativity constraint is violated.

The work does well in setting up the mutual information maximization under PSL constraints and providing a transceiver design plus approximate BER analysis for multipath channels. The numerical comparisons to generalized PAS are presented as showing better PSLR control and weak-target detection.

The soft spots are around the load-bearing assumption. The stress-test concern holds up from the abstract: the exact control only applies when the spectral non-negativity is satisfied, yet the paper does not appear to demonstrate that this happens with high probability for the proposed construction. Without that, the advantage reduces to the weaker probabilistic bound, and it's not obvious how much better that is in practice. The channel model is also restricted to quasi-static frequency-flat for the optimality claim.

This is aimed at researchers focused on waveform design for integrated sensing and communications. Anyone working on 6G ISAC signals would find the specific construction and the PSL bounds useful to consider.

I would send it for peer review. The idea is concrete enough and the analysis is grounded enough to warrant referee input, even with the open questions on the constraint's frequency.

Referee Report

2 major / 1 minor

Summary. The paper proposes auto-correlation function keying (ACFK) for communication-centric ISAC. It formulates mutual-information maximization under peak sidelobe level (PSL) constraints and power budget. For quasi-static frequency-flat channels, a continuous ACF-domain uniform construction is shown to be asymptotically optimal at high SNR. ACFK embeds data symbols directly onto ACF-domain sidelobes to achieve exact control of the nominal periodic auto-correlation function (P-ACF), which coincides with the actual P-ACF when a spectral non-negativity constraint holds; otherwise the violation probability is quantified and the resulting PSLR degradation is bounded. A reference transceiver design is given for quasi-static multipath channels together with high-SNR approximate BER analysis. Numerical comparisons against a generalized probabilistic amplitude shaping (PAS) baseline show stronger PSLR control and improved weak-target detection.

Significance. If the central claims hold, the work supplies a concrete mechanism for controlling per-realization peak sidelobes in random payload signals rather than only expected sidelobe levels, directly addressing a limitation in existing ISAC sensing analyses and offering measurable gains in weak-target detection under comparable rate and power constraints.

major comments (2)

[Abstract] Abstract: the headline claim of 'exact control of the nominal P-ACF, which coincides with the actual P-ACF' is load-bearing for the reported performance advantage over generalized PAS, yet the manuscript only quantifies the non-negativity violation probability without demonstrating that this probability is vanishing (or sufficiently close to zero) for the ACF-domain uniform construction; if the violation rate remains non-vanishing, the exact-control guarantee reduces to the weaker probabilistic PSLR bound and the claimed 'substantially stronger PSLR control' is no longer supported.
[Abstract] Abstract: the asymptotic optimality result is derived under the explicit assumption of quasi-static frequency-flat channels at high SNR; the finite-constellation ACFK inherits the same spectral non-negativity requirement, but no explicit verification is provided that the optimality principle continues to hold once the continuous construction is discretized and the non-negativity constraint is enforced only probabilistically.

minor comments (1)

[Abstract] The abstract states that 'high-SNR approximate BER analysis' is provided but does not indicate the order of the approximation or the error term that is neglected.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on the abstract claims regarding exact P-ACF control and asymptotic optimality. We address both major comments point-by-point below with clarifications and proposed revisions to strengthen the manuscript.

read point-by-point responses

Referee: [Abstract] Abstract: the headline claim of 'exact control of the nominal P-ACF, which coincides with the actual P-ACF' is load-bearing for the reported performance advantage over generalized PAS, yet the manuscript only quantifies the non-negativity violation probability without demonstrating that this probability is vanishing (or sufficiently close to zero) for the ACF-domain uniform construction; if the violation rate remains non-vanishing, the exact-control guarantee reduces to the weaker probabilistic PSLR bound and the claimed 'substantially stronger PSLR control' is no longer supported.

Authors: We acknowledge the concern that stronger evidence of low violation probability would better support the 'substantially stronger PSLR control' claim. The manuscript already quantifies the non-negativity violation probability for the ACF-domain uniform construction and derives a bound on the resulting PSLR degradation. To directly address this point, we will add numerical evaluations in the revised results section demonstrating that the violation probability is small (e.g., below 0.01 for practical parameters at high SNR). This evidence will confirm that the nominal and actual P-ACF coincide with high probability, thereby preserving the exact-control advantage. We will also refine the abstract wording to more explicitly note the high-probability nature of the coincidence. revision: yes
Referee: [Abstract] Abstract: the asymptotic optimality result is derived under the explicit assumption of quasi-static frequency-flat channels at high SNR; the finite-constellation ACFK inherits the same spectral non-negativity requirement, but no explicit verification is provided that the optimality principle continues to hold once the continuous construction is discretized and the non-negativity constraint is enforced only probabilistically.

Authors: The asymptotic optimality result is presented strictly as the motivating design principle for the continuous ACF-domain uniform construction under the stated channel and SNR assumptions. ACFK discretizes this construction while handling the non-negativity constraint probabilistically, as analyzed in the paper. The high-SNR approximate BER analysis and numerical comparisons already show that the performance benefits extend to the finite case. We agree an explicit link would strengthen the presentation; we will add a short remark in the discussion of the design principle (Section III) explaining that discretization error vanishes asymptotically at high SNR and the probabilistic enforcement has bounded impact per the existing PSLR degradation bound. This clarifies the extension without changing the core claims. revision: partial

Circularity Check

0 steps flagged

No significant circularity; derivation is self-contained from stated optimization.

full rationale

The central claim derives the continuous ACF-domain uniform construction as asymptotically optimal directly from the mutual information maximization problem under PSL constraints for quasi-static frequency-flat channels at high SNR. ACFK is motivated by this principle and provides nominal P-ACF control, with explicit quantification of non-negativity violation probability and PSLR degradation bounds when the constraint fails. All performance claims are benchmarked against an external generalized PAS baseline rather than internal fitted quantities. No self-definitional reductions, fitted inputs renamed as predictions, or load-bearing self-citations appear in the derivation chain.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 1 invented entities

The paper builds on standard mutual information maximization and correlation-function analysis from information theory and signal processing. The primary addition is the ACFK construction itself under conventional ISAC channel models.

axioms (2)

domain assumption Quasi-static frequency-flat channels allow the continuous ACF-domain uniform construction to be asymptotically optimal at high SNR
Invoked to establish the high-SNR design principle for the mutual information maximization problem.
domain assumption Spectral non-negativity constraint ensures nominal P-ACF coincides with actual P-ACF
Required for the claim of exact P-ACF control; violation probability is quantified separately.

invented entities (1)

ACFK modulation architecture no independent evidence
purpose: Embed data symbols directly onto ACF-domain sidelobes to enforce PSL constraints
Newly proposed finite-constellation scheme motivated by the continuous uniform construction.

pith-pipeline@v0.9.1-grok · 5841 in / 1592 out tokens · 43033 ms · 2026-06-26T22:37:32.893502+00:00 · methodology

0 comments

read the original abstract

We propose ACFK: Auto-correlation Function Keying, a new integrated sensing and communication (ISAC) waveform that carries random communication data while directly controlling the peak sidelobe level (PSL) of the periodic auto-correlation function (P-ACF). In contrast to existing works aiming at controlling the expected sidelobe level (ESL), which fails to characterize realization-specific sidelobe behaviors, we formulate a mutual information maximization problem under PSL and power constraints, and show that a continuous ACF-domain uniform distribution is asymptotically optimal at high signal-to-noise ratio (SNR) over quasi-static frequency-flat channels. Motivated by this principle, ACFK maps finite-constellation symbols onto auto-correlation function (ACF)-domain sidelobes and uses independent phase symbols to exploit the remaining degrees of freedom. The resulting waveform enables exact control of the nominal P-ACF, which coincides with the actual P-ACF when the power spectral non-negativity condition is satisfied. We further analyze the non-negativity violation probability and bound the corresponding peak sidelobe level ratio (PSLR) degradation. A reference ISAC transceiver and its high-SNR approximate bit error rate (BER) analysis are also provided. Numerical results show that ACFK achieves stronger PSLR control, and improved weak-target detection performance, than a generalized probabilistic amplitude shaping (PAS) baseline at similar data rate and BER.