Fourier-Domain CFO Estimation Using Jutted Binary Modulation on Conjugate-Reciprocal Zeros

Alphan Sahin; Anthony Joseph Perre; Parker Huggins

REVIEW 2 major objections 7 minor 9 references

Fourier-Domain CFO Estimation Using Jutted Binary Modulation on Conjugate-Reciprocal Zeros

T0 review · 2 major / 7 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read By moving one zero outward in the BMOCZ constellation, a receiver can estimate the whole carrier frequency offset from the received spectrum alone, removing the need for pilots or cyclically permutable codes.

desk verdict A clever twist on BMOCZ with a fixable math error in the AACF definition; worth reviewing. read the letter →

arxiv 2506.07256 v2 pith:QLAMKWGP submitted 2025-06-08 eess.SP

Parker Huggins , Anthony Joseph Perre , Alphan Sahin This is my paper · ORCID

classification eess.SP

keywords carrierfrequencyoffset(CFO)binarymodulationonconjugate-reciprocalzeros(BMOCZ)HuffmanBMOCZaperiodicautocorrelationfunctionIDFT-basedCFOestimationnon-coherentcommunicationzeropilot-freesynchronization

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 claims that a small, deliberate asymmetry in the zero constellation of BMOCZ — moving one zero outward along the real axis — lets a receiver estimate a carrier frequency offset anywhere in $[0, 2\pi)$ without pilots and without channel coding. The reason the trick works is that every BMOCZ codeword shares the same aperiodic autocorrelation function, so the receiver knows the magnitude spectrum template in advance; a CFO simply rotates all zeros and shifts that template, and the receiver finds the shift by an IDFT-based correlation. This matters because existing Huffman BMOCZ CFO correction forces a cyclically permutable code, which fixes the coding scheme and constrains message length. If the claim holds, the constellation shape alone resolves the CFO ambiguity, at a cost of roughly 1 dB of BER relative to Huffman BMOCZ with no CFO in AWGN, and it opens BMOCZ to standard codes and soft-decision decoding.

What carries the argument

The load-bearing object is the identical aperiodic autocorrelation function $A(z)$, defined for the BMOCZ sequence $\mathbf{x}$ and expressible from its zeros as $X(z)X(1/z)$; evaluated on the unit circle it should equal $|X(e^{j\omega})|^2$, which makes the template vector $\mathbf{t}_N$ (the magnitude of the $N$-point IDFT of a codeword) known to the receiver independent of the message. The 'jutted' zero — radius $\zeta R$ for $k=0$, radius $R$ for the other $K-1$ zeros — is the second piece: it breaks the $K$-fold rotational symmetry of the Huffman constellation so the template is non-periodic and the correlation has a unique peak. Algorithm 1 carries the estimation: it generates a modulation matrix $\mathbf{M}$ of candidate rotations, forms $N$ modulated copies of the received sequence, takes their $N$-point IDFTs, and picks the rotation whose magnitude column has the largest inner product with $\mathbf{t}_N$, then refines the search interval over iterations.

What would settle it

Run Algorithm 1 on a simulated received signal with a known CFO $\varphi_0$ and a two-tap channel ($L_e = 2$). If the inner-product peak in Eq. (13) shifts away from $\varphi_0$ or develops multiple comparable peaks, the pilot-free claim is confined to the flat-fading case. Separately, evaluate Eq. (5) with $X(z)\overline{X(1/\bar{z})}$ and compare it to $|X(e^{j\omega})|^2$ for the jutted radii: the equality in Eq. (9) is load-bearing, and any numerical mismatch changes the template $\mathbf{t}_N$ and the CFO estimate.

Watch

Extended reading notes

Core claim

The paper's central discovery is that the CFO ambiguity of Huffman BMOCZ is an artifact of rotational symmetry, not of non-coherent zero modulation itself. In Huffman BMOCZ the zeros sit on two concentric circles, so any rotation by the base angle $2\pi/K$ maps a valid message to another valid message. J-BMOCZ breaks that symmetry by setting the radius of the first zero pair to $\zeta R$ instead of $R$, creating a 'jutted' zero while leaving all other zeros at the Huffman radii. Because the AACF — and hence the magnitude spectrum — of BMOCZ sequences is message-independent, the receiver can build the template $\mathbf{t}_N$ and estimate the full CFO $\varphi$ by aligning $\mathbf{t}_N$ with the magnitude spectrum of the received sequence through the modulation matrix and inner-product maximization in Eq. (13). The paper reports that uncoded J-BMOCZ under a CFO loses about 1 dB BER over CFO-free Huffman BMOCZ in AWGN (about 2 dB in fading), and that J-BMOCZ with a (31,16)-BCH code outperforms Huffman BMOCZ with an ACPC under CFO by about 1.75 dB in BER in fading, while keeping similar BLER.

Load-bearing premise

The estimator assumes every codeword shares a single known magnitude-spectrum template — the identical AACF — and that the channel contributes no zeros ($L_e = 1$), so the received spectrum is the template shifted by the CFO.

Editorial extensions

If this is right

Pilot-free and channel-coding-free CFO correction becomes possible for BMOCZ in flat-fading links, releasing the message length and coding constraints imposed by cyclically permutable codes.
Uncoded J-BMOCZ keeps a roughly 1 dB BER loss relative to CFO-free Huffman BMOCZ in AWGN under a CFO uniformly drawn from $[0, 2\pi)$, with the loss growing to about 2 dB in a fading channel.
Coded J-BMOCZ with a (31,16)-BCH code beats Huffman BMOCZ with the (31,16)-ACPC by about 1.75 dB in BER under CFO in fading, at similar BLER, while also avoiding the ACPC's fixed outer code.
Because no cyclic permutability constraint is needed, J-BMOCZ is compatible with soft-decision decoding and standard polar or LDPC codes, which the paper identifies as a likely performance improvement.
The iterative search with $N=64$, $\delta=0.2$, and two iterations resolves the full CFO range, so the added receiver complexity over the baseline DiZeT decoder is a modest number of small IDFTs.

Reading between the lines

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

Editorial inference: the template correlation should remain partially functional for short channels with more than one tap when the channel zeros sit away from the unit circle, since they add ripple to $|Y(e^{j\omega})|$ rather than a clean shift; this can be tested by running Algorithm 1 on a two-tap channel with known zeros.
Editorial inference: $\zeta$ is a tunable trade-off — larger $\zeta$ sharpens the template peak and the CFO estimate but should deepen the baseline BER penalty — so an analytic $\zeta(K)$ rule, which the paper leaves to future work, would let operators choose the operating point in advance.
Editorial inference: the same correlation logic applies to any message-independent spectral feature, so variants of J-BMOCZ with multiple jutted zeros could trade a larger template peak for more spectrum-shaping flexibility.
Editorial inference: the no-interference assumption means the practical regime is small CFOs relative to bandwidth, as the paper itself notes; an OFDM implementation could equally use this estimator for a timing offset, which would make it a joint synchronization primitive rather than only a CFO fix.

Share X Bluesky LinkedIn Reddit HN

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

The new idea is simple and clever: jut one zero out radially to break Huffman's rotational symmetry, then estimate CFO by correlating the received magnitude spectrum against a message-independent template. This removes the need for cyclically permutable codes and opens the door to standard channel codes, a real benefit for this niche. The worked K=2 example checks out, and the simulation results (1 dB loss for uncoded J-BMOCZ under CFO vs Huffman without CFO; coded J-BMOCZ beating BMOCZ-ACPC) are plausible.

The soft spots: first, the AACF in Definition 1 (Eq. 4) is written without conjugation, but Eq. 9 evaluates it as |X|^2, which corresponds to the conjugate correlation. The zero-domain formula in Eq. 5 is accordingly wrong for complex zeros: it gives zeros at alpha and 1/alpha, not alpha and 1/bar{alpha}. For the uniform-phase Huffman/J-BMOCZ constellations with K >= 4, most zeros are off the real axis, so the identity 'all sequences share the same AACF' does not follow from the printed equations. The K=2 example uses real zeros only, so it cannot catch the error. The corrected identity A(z)=X(z)X^*(1/z^*) is standard and true for conjugate-reciprocal zeros, so I expect the estimator is fine, but the paper needs to fix the notation and re-derive.

Second, zeta is selected by sweeping to minimize BER on the same metric report; that's a tuned parameter, and without cross-validation or an analysis of sensitivity it's hard to know how robust the 1 dB claim is. Also, no code or data are shared, and the flat-fading assumption (Le=1) is assumed throughout, so the claim that it works in fading rests on that idealization. These are presentation and reproducibility concerns, not fatal flaws.

Bottom line: this is a solid contribution to a niche subfield, the math just needs a correction. I'd send it to review, with a request to fix the AACF definition, report the zeta sweep details, and release simulation code. It's worth a serious referee.

Referee Report

2 major / 7 minor

Summary. The paper proposes J-BMOCZ, a modification of Huffman BMOCZ in which the zero associated with the first message bit is placed at radius ζR (or its reciprocal) instead of the common radius R, thereby breaking the rotational symmetry that makes the integer CFO ambiguous in Huffman BMOCZ. It then derives a pilot-free CFO estimator: because BMOCZ codewords share a message-independent magnitude spectrum |X(e^{jω})|, the receiver cross-correlates the magnitude spectrum of the received block with this known template over candidate rotations, using an iterative IDFT-based search, and then applies standard DiZeT decoding. The paper includes a fully worked K=2 example and validates the method by BER and BLER simulations in AWGN and flat fading, reporting a roughly 1 dB loss in AWGN (2 dB in fading) for uncoded J-BMOCZ under CFO relative to Huffman BMOCZ without CFO, and better coded BER than Huffman BMOCZ with an ACPC under CFO.

Significance. If the corrected derivation holds, the paper removes a real limitation of Huffman BMOCZ: CFO estimation without a cyclically permutable code, so arbitrary channel codes and message lengths can be used in conjunction with the non-coherent DiZeT decoder. The approach is simple, the algorithm is explicit, and the K=2 example is arithmetically checkable; the equal-rate comparison between coded J-BMOCZ and (31,16)-ACPC Huffman BMOCZ is a fair benchmark. These strengths support the paper's interest for short-packet non-coherent communication. The main caveats are that the central proof as printed (Eqs. (4)-(5)) is internally inconsistent and must be restated with the conjugate AACF, and that the headline 1 dB figure is obtained at a ζ tuned on the reported BER metric, so sensitivity to ζ should be documented.

major comments (2)

[Section II-A, Def. 1, Eqs. (4)-(5); Section III-A, Eq. (9)] The AACF in Definition 1 is defined without conjugation (a_ℓ = Σ_i x_i x_{i+ℓ}), and Eq. (5) gives A(z)=X(z)X(1/z), but Eq. (9) evaluates A(e^{jω}) as |X(e^{jω})|^2, which corresponds to the conjugate correlation A(z)=X(z)\overline{X(1/\bar{z})}. The un-conjugated product in Eq. (5) is message-dependent whenever any ψ_k is not 0 or π: the two bit choices for zero k produce the root sets {r_k e^{jψ_k}, r_k^{-1} e^{-jψ_k}} and {r_k^{-1} e^{jψ_k}, r_k e^{-jψ_k}}, which coincide only for real zeros. For J-BMOCZ with K≥4, ψ_k=2πk/K includes complex phases, so the printed derivation does not establish that all codewords share one magnitude template. The claim is nevertheless correct after the standard fix a_ℓ=Σ_i x_i\overline{x}_{i+ℓ} (equivalently A(z)=X(z)\overline{X(1/\bar{z})}), under which the conjugate-reciprocal pair {α_k, 1/\bar{α}_k} is independent of b_k; the authors should correct Definition 1 and Eq. (5) accordingly and state that the simulations use this conjugate form. The only worked example (K=2, ψ_k ∈ {0,π}) has real zeros and therefore cannot expose the inconsistency. Relatedly, the parenthetical after Eq. (1) states (r_k e^{jψ_k})^{-1}=r_k^{-1}e^{jψ_k}; this is the ordinary reciprocal only if r_k e^{jψ_k} is real, and the intended conjugate-reciprocal identity requires conjugation of r_k e^{jψ_k} before inversion.
[Section IV, parameter selection for ζ] The asymmetry factor is chosen by sweeping its value and selecting that which minimizes the BER under a CFO, and the reported 1 dB (AWGN) and 2 dB (fading) losses, as well as the coded comparisons, are all evaluated at the resulting ζ=1.15. Because ζ is tuned on the same BER metric that is then reported, the quantitative claims are conditional on an in-sample optimum. The authors should report BER versus ζ (e.g., at a representative Eb/N0), state the sensitivity of the 1 dB figure to ζ and to K, and indicate whether ζ=1.15 is fixed across SNR, channel, and block length or re-tuned per simulation.

minor comments (7)

[Abstract vs. Section IV-A] The abstract states that the BER loss under CFO is just 1 dB, but Section IV-A reports a 2 dB loss in the fading channel; the abstract should qualify the figure as the AWGN result.
[Section III-A, Eqs. (10) and (12)] The manuscript calls the transform used for t_N and for the columns of \tilde{Y}_N the 'N-point IDFT'; with the standard DFT/IDFT convention these are N-point DFT evaluations of the zero-padded sequence. Please define the transform convention once so that the sign in Eq. (12) is unambiguous.
[Section III-A, sentence after Eq. (10)] The sentence 't_N is equivalent to the magnitude of the N-point IDFT of x, which is known at the receiver' conflates the unknown transmitted codeword with the known template; rephrase to say the template is known because, by the corrected AACF property, |X(e^{jω})| is independent of the message.
[Section III-IV] The manuscript does not report CFO estimation error (e.g., RMSE of \hat{φ} versus Eb/N0, N, or ζ); such a curve would directly validate the Section III claim that the CFO is estimable over [0,2π).
[Eq. (6), DiZeT scaling] The DiZeT rule in Eq. (6) applies the per-zero scaling r_k^K; for J-BMOCZ r_k differs for k=0, so state explicitly that the scaling uses r_k from (1)/(14) and note that this is the natural generalization of the Huffman rule.
[Section IV, simulation details] Please state the number of Monte Carlo realizations used for the BER/BLER curves in Figs. 4-5, or add confidence intervals, so that the reported 1 dB and fractional-dB differences are meaningful.
[Algorithm 1] The refinement window half-width in iteration ι is δ/ι; for the true CFO to remain in the window after the coarse pass, δ should exceed the coarse bin width 2π/N (here δ=0.2 with N=64 does so). State this condition or comment on the lock-loss behavior.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the J-BMOCZ CFO estimator uses the known, message-independent BMOCZ magnitude template, and no fitted parameter is disguised as a prediction.

full rationale

The CFO estimator in Section III-A is not circular. The template t_N in Eq. (10) is defined as |X(e^{jω})| for a generic codeword and is known to the receiver because all BMOCZ/J-BMOCZ codewords share the same AACF, hence the same magnitude spectrum. This is an external, message-independent constellation property from prior work [2], not a function of the received signal or of a fitted parameter. The estimate in Eq. (13) maximizes the cross-correlation between this fixed template and the IDFT magnitude of the CFO-rotated received sequence; no estimated quantity is used to construct the template. The asymmetry factor ζ in Eq. (14) is selected offline by sweeping to minimize BER (Section IV) and its performance is then reported; this is standard parameter tuning, not a fitted input renamed as a prediction. The only substantive caveat is a proof-consistency issue, not circularity: Definition 1 / Eq. (4) defines the AACF without conjugation, while Eq. (9) uses the conjugate form |X|^2, and the zero-domain identity in Eq. (5) as printed (X(z)X(1/z)) is only message-independent for real zeros. The conjugate-reciprocal BMOCZ property requires X(z)X^*(1/z^*), which makes the printed derivation of message-independence incomplete. However, the underlying property is true and externally established, and the estimator does not reduce to its own inputs. Therefore no circular step rises to the level of the seven enumerated patterns.

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

The central claim rests on the message-independent AACF and template, the flat-fading channel model, and the CFO-as-phase-rotation model. The only fitted number is zeta, selected by simulation sweep; algorithm constants N, delta, and niterations are hand-chosen. No new physical entities are introduced.

free parameters (4)

zeta (asymmetry factor) = 1.15 (for K=31 and K=32)
Chosen by sweeping over values to minimize BER under CFO; this is a design parameter tuned to the performance metric, not derived.
N (number of frequency samples) = 64
Number of frequency samples and search grid points; hand-chosen algorithm parameter.
delta (refinement window) = 0.2
Iterative refinement window size; hand-chosen algorithm parameter.
niterations = 2
Number of refinement iterations; hand-chosen algorithm parameter.

assumptions (4)

domain assumption All BMOCZ codewords share an identical aperiodic autocorrelation function (AACF), so the magnitude spectrum template T(omega) is message-independent.
Used in Section III-A to construct t_N; the identity follows from conjugate-reciprocal zero pairing, but the paper asserts it without proof and relies on prior work [2].
domain assumption The channel is flat-fading with effective length Le=1, so Y(z)=hX(z)+W(z) and the received magnitude spectrum is a scaled, shifted version of |X(e^{j omega})|.
Section II states this assumption; if Le>1, channel zeros appear and the template-correlation CFO estimate breaks down.
domain assumption A CFO appears as a linear phase modulation m_phi on the received sequence with zero intersymbol interference.
Section II-B Eq. (7)-(8); the paper assumes CFO is bounded so that no ISI occurs.
domain assumption The template vector can be computed at the receiver from the known AACF without knowing the message.
Section III-A states t_N is known at the receiver; this presumes the AACF is available to the receiver as public side information.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Fourier-Domain CFO Estimation Using Jutted Binary Modulation on Conjugate-Reciprocal Zeros." pith.science (2026). https://pith.science/paper/QLAMKWGP

@misc{pith2026250607256,
  author       = {Pith},
  title        = {Pith review of: Fourier-Domain CFO Estimation Using Jutted Binary Modulation on Conjugate-Reciprocal Zeros},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QLAMKWGP}},
  note         = {Machine review of arXiv:2506.07256}
}

read the original abstract

In this work, we propose jutted binary modulation on conjugate-reciprocal zeros (J-BMOCZ) for non-coherent communication under a carrier frequency offset (CFO). By introducing asymmetry to the Huffman BMOCZ zero constellation, we exploit the identical aperiodic auto-correlation function of BMOCZ sequences to derive a Fourier-domain metric for CFO estimation. Unlike the existing methods for Huffman BMOCZ, which require a cyclically permutable code (CPC) for pilot-free CFO correction, J-BMOCZ enables the estimation of a CFO without the use of pilots or channel coding. Through numerical simulations in additive white Gaussian noise and fading channels, we show that the bit error rate (BER) loss of J-BMOCZ under a CFO is just 1 dB over Huffman BMOCZ without a CFO. Furthermore, the results show that coded J-BMOCZ achieves better BER performance than Huffman BMOCZ with a CPC.

Figures

Figures reproduced from arXiv: 2506.07256 by the authors.

**Figure 1.** Example Huffman BMOCZ zero pattern for K = 8 and R = 1.176. (a) Full zero constellation with 2K zero positions. (b) Transmitted zeros corresponding to the message b = (1, 0, 1, 1, 1, 0, 0, 1). (c) Received zeros rotated by φ = (12/7)θK radians due to a CFO. (d) Received zeros after correcting ǫ, which correspond to a cyclic permutation of the transmitted message, i.e., bˆ = (1, 1, 0, 1, 1, 1, 0, 0). Hence, in the pr… view at source ↗

**Figure 2.** shows the template vector tN for J-BMOCZ with K = 16 and ζ ∈ {1, 1.1, 1.2}. Notice that for ζ > 1, a peak is introduced to tN that breaks the periodicity of the sinusoidal Huffman BMOCZ template vector. The J-BMOCZ template vector is thus well-suited for CFO estimation using the approach of Section III-A, as the inner product in (13) is -100 -50 0 50 100 n 2 3 4 5 6 7 8 t n = 1.0 (Huffman BMOCZ) = 1.1 (J-BMOCZ) = 1.… view at source ↗

**Figure 3.** Proposed J-BMOCZ zero pattern for K = 8, R = 1.176, and ζ = 1.15. (a) Full zero constellation with 2K zero positions. (b) Transmitted zeros corresponding to the message b = (1, 0, 1, 1, 1, 0, 0, 1). (c) Received zeros rotated by φ = (12/7)θK radians due to a CFO. Received zeros after correcting φ using Algorithm 1, which exactly correspond to the transmitted message. K + 1. The transmitted polynomials are perturbed … view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: Comparison of uncoded J-BMOCZ to uncoded Huffman BMO [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]

**Figure 5.** Figure 5: Comparison of coded J-BMOCZ to coded Huffman BMOCZ. B [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

9 extracted references · 7 canonical work pages

[1]

Short-message communi cation and FIR system identiﬁcation using Huffman sequences,

P . Walk, P . Jung, and B. Hassibi, “Short-message communi cation and FIR system identiﬁcation using Huffman sequences,” in Proc. IEEE International Symposium on Information Theory (ISIT) , 2017, pp. 968– 972

work page 2017
[2]

MOCZ for blind short-packet communication: Basic p rinciples,

——, “MOCZ for blind short-packet communication: Basic p rinciples,” IEEE Transactions on Wireless Communications , vol. 18, no. 11, pp. 5080–5097, 2019

work page 2019
[3]

MOCZ for blind short- packet communication: Practical aspects,

P . Walk, P . Jung, B. Hassibi, and H. Jafarkhani, “MOCZ for blind short- packet communication: Practical aspects,” IEEE Transactions on Wireless Communications, vol. 19, no. 10, pp. 6675–6692, 2020

work page 2020
[4]

The perﬁdious polynomial,

J. H. Wilkinson, “The perﬁdious polynomial,” Studies in Numerical Analysis, vol. 24, pp. 1–28, 1984

work page 1984
[5]

Integrated sensing and communication with MOCZ w aveform,

S. K. Dehkordi, P . Jung, P . Walk, D. Wieruch, K. Heuermann , and G. Caire, “Integrated sensing and communication with MOCZ w aveform,” arXiv preprint arXiv:2307.01760 , 2023

arXiv 2023
[6]

Over-the-air majority vote computation wit h modulation on conjugate-reciprocal zeros,

A. S ¸ ahin, “Over-the-air majority vote computation wit h modulation on conjugate-reciprocal zeros,” IEEE Transactions on Wireless Communica- tions, vol. 23, no. 11, pp. 17 714–17 726, 2024

work page 2024
[7]

Non-coherent short-packet communications: N ovel z-domain user multiplexing,

T. Eren, “Non-coherent short-packet communications: N ovel z-domain user multiplexing,” Digital Signal Processing , vol. 156, p. 104777, 2025

work page 2025
[8]

On the optimal radius and subca rrier mapping for binary modulation on conjugate-reciprocal zeros,

P . Huggins and A. S ¸ ahin, “On the optimal radius and subca rrier mapping for binary modulation on conjugate-reciprocal zeros,” in Proc. IEEE Military Communications Conference (MILCOM) , 2024, pp. 1–6

2024

Show all 9 references

[9]

The generation of impulse-equivalent puls e trains,

D. Huffman, “The generation of impulse-equivalent puls e trains,” IRE Transactions on Information Theory , vol. 8, no. 5, pp. 10–16, 1962

1962

Pith tools

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

[1] [1]

Short-message communi cation and FIR system identiﬁcation using Huffman sequences,

P . Walk, P . Jung, and B. Hassibi, “Short-message communi cation and FIR system identiﬁcation using Huffman sequences,” in Proc. IEEE International Symposium on Information Theory (ISIT) , 2017, pp. 968– 972

work page 2017

[2] [2]

MOCZ for blind short-packet communication: Basic p rinciples,

——, “MOCZ for blind short-packet communication: Basic p rinciples,” IEEE Transactions on Wireless Communications , vol. 18, no. 11, pp. 5080–5097, 2019

work page 2019

[3] [3]

MOCZ for blind short- packet communication: Practical aspects,

P . Walk, P . Jung, B. Hassibi, and H. Jafarkhani, “MOCZ for blind short- packet communication: Practical aspects,” IEEE Transactions on Wireless Communications, vol. 19, no. 10, pp. 6675–6692, 2020

work page 2020

[4] [4]

The perﬁdious polynomial,

J. H. Wilkinson, “The perﬁdious polynomial,” Studies in Numerical Analysis, vol. 24, pp. 1–28, 1984

work page 1984

[5] [5]

Integrated sensing and communication with MOCZ w aveform,

S. K. Dehkordi, P . Jung, P . Walk, D. Wieruch, K. Heuermann , and G. Caire, “Integrated sensing and communication with MOCZ w aveform,” arXiv preprint arXiv:2307.01760 , 2023

arXiv 2023

[6] [6]

Over-the-air majority vote computation wit h modulation on conjugate-reciprocal zeros,

A. S ¸ ahin, “Over-the-air majority vote computation wit h modulation on conjugate-reciprocal zeros,” IEEE Transactions on Wireless Communica- tions, vol. 23, no. 11, pp. 17 714–17 726, 2024

work page 2024

[7] [7]

Non-coherent short-packet communications: N ovel z-domain user multiplexing,

T. Eren, “Non-coherent short-packet communications: N ovel z-domain user multiplexing,” Digital Signal Processing , vol. 156, p. 104777, 2025

work page 2025

[8] [8]

On the optimal radius and subca rrier mapping for binary modulation on conjugate-reciprocal zeros,

P . Huggins and A. S ¸ ahin, “On the optimal radius and subca rrier mapping for binary modulation on conjugate-reciprocal zeros,” in Proc. IEEE Military Communications Conference (MILCOM) , 2024, pp. 1–6

2024