Chirp-Permuted AFDM for Quantum-Resilient Physical-Layer Secure Communications

Giuseppe Thadeu Freitas de Abreu; Hyeon Seok Rou

REVIEW 3 major objections 4 minor 21 references

Chirp-Permuted AFDM for Quantum-Resilient Physical-Layer Secure Communications

T0 review · 3 major / 4 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read The paper claims that keeping only the permutation order of the second chirp sequence secret in an AFDM waveform makes transmissions undecodable to eavesdroppers, even ones with perfect channel knowledge and co-location.

desk verdict Clean combinatorial core, but the 'virtually perfect security' claim fails against a known-pilot attack that recovers the permutation directly. read the letter →

arxiv 2502.03289 v2 pith:6RMPHDTZ submitted 2025-02-05 eess.SP

Hyeon Seok Rou , Giuseppe Thadeu Freitas de Abreu This is my paper · ORCID

classification eess.SP

keywords physical-layersecurityAFDMchirp-permutationquantumresiliencewirelesscommunicationsbeyond5GGroveradaptivesearchderangements

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 proposes a physical-layer security scheme built on chirp-permuted AFDM, in which the transmitter and legitimate receiver share a single secret: the permutation order of the waveform's second chirp sequence. It claims that this secret alone suffices, because an eavesdropper who does not know the permutation cannot decode the data, even with perfect channel state information and even when co-located with the legitimate receiver and seeing the identical signal. The security argument has two prongs: the key space contains N! permutations, so exhaustive search costs O(λN!) classically and O(√N!) queries under Grover adaptive search, and the probability of a blind guess landing on or near the true permutation is governed by derangement statistics and becomes negligible. Simulations for N=64 show the legitimate receiver's bit-error rate falling with SNR while the eavesdropper's stays near 0.5 in both remote and co-located scenarios, including when the guessed permutation has up to ten correctly placed elements. If the claim holds, this would provide quantum-resilient physical-layer security without extra energy, hardware, or knowledge of the eavesdropper's channel.

What carries the argument

The load-bearing object is the second chirp sequence $\lambda_{c_2}$ of the AFDM transform, a vector of quadratic phase factors $e^{-j2\pi c_2 n^2}$ for $n = 0, \ldots, N-1$, permuted by the secret order $k$ into the diagonal matrix $\Lambda_{c_2,k}$. This matrix enters the DAFT as $A_k = \Lambda_{c_2,k} F_N \Lambda_{c_1}$; the transmitter applies $A_k^{-1}$ and the legitimate receiver applies $A_k$. Since both chirp operations are diagonal, the effective channel $G_k = A_k H A_k^{-1}$ is a similarity transform of an intermediate channel, preserving the positions of nonzero delay-Doppler coefficients for the matched receiver while changing the coefficients completely for a mismatched permutation. The security apparatus then counts permutations: the $N!$ key space, derangement probabilities for near-guesses, and the Grover adaptive search query complexity $O(\sqrt{N!})$ attributed to a quantum eavesdropper.

What would settle it

Transmit a known pilot symbol, have an eavesdropper with perfect channel knowledge compute $t = \Lambda_{c_1} F_N H^{-1} r$ and divide by the pilot; if the phases of $t$ reveal $e^{j2\pi c_2 \sigma(j)^2}$, the permutation $\sigma$ is recovered in polynomial time, refuting the undecodability claim. A second check is to set $c_2 = a/N$ with integer $a$ and count the distinct matrices $\Lambda_{c_2,i}$; if that count falls below $N!$, the factorial-complexity argument no longer applies.

Watch

Extended reading notes

Core claim

At the center of the scheme is the discrete affine Fourier transform (DAFT) of AFDM, written $A = \Lambda_{c_2} F_N \Lambda_{c_1}$, where $\Lambda_{c_1}$ and $\Lambda_{c_2}$ are diagonal chirp matrices whose entries are quadratic phase factors $e^{-j2\pi c n^2}$. The proposal permutes the entries of the second chirp according to a shared secret order $k$, giving $A_k = \Lambda_{c_2,k} F_N \Lambda_{c_1}$ and transmitted signals $s_k = A_k^{-1} x$. Because the chirp operations are diagonal, demodulation with the correct $A_k$ preserves the delay-Doppler sparsity structure of the effective channel, while demodulation with any wrong permutation produces completely different channel coefficients. The paper proves by combinatorial analysis that the probability of a random permutation containing exactly $\ell$ correctly placed entries is $P_\ell = D_{N-\ell}/N! = (1/\ell!) \sum_{n=0}^{N-\ell} (-1)^n/n!$, so for large $N$ a guess with more than about ten correct entries is virtually impossible, and BER simulations show that even ten correct entries leave the eavesdropper at chance level. The claim is that the permutation order alone is sufficient for virtually perfect physical-layer security, resistant to both classical and quantum-accelerated search.

Load-bearing premise

The load-bearing premise is that the eavesdropper's only effective moves are exhaustive search or blind guessing of the permutation, and that all N! permutations yield distinct chirp sequences; if a known reference symbol lets the eavesdropper read the permutation from received phases, or if repeated chirp entries make different keys equivalent, the claimed security guarantee breaks.

Editorial extensions

If this is right

For a 5G-scale system with N=3300 subcarriers, the N! key space makes classical exhaustive search infeasible and a Grover-based quantum search would need on the order of 10^1654 queries and roughly 10^7-10^8 error-corrected qubits, beyond projected 2040 hardware.
A blind guess of the permutation succeeds with probability at most 1/N!, and for large N a random guess has a 99.999999992% chance of placing no more than ten elements correctly.
With a wrong permutation, MMSE or ML detection leaves the eavesdropper's bit-error rate near 0.5, so no data bits are recovered in either remote or co-located interception scenarios.
The legitimate link retains AFDM's spectral efficiency and delay-Doppler diversity, because the chirp permutation acts as a similarity transformation on the effective channel rather than changing its sparsity structure.
The scheme needs no artificial noise, no extra hardware, and no transmitter knowledge of the eavesdropper's channel; the shared secret is simply the permutation order.

Reading between the lines

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

An attack not analyzed in the paper: coherent receivers rely on known pilots, and with perfect CSI an eavesdropper could compute $t = \Lambda_{c_1} F_N H^{-1} r$, divide by the pilot, and read the phase pattern $e^{j2\pi c_2 \sigma(j)^2}$, recovering the permutation $\sigma$ in polynomial time.
If the chirp parameter $c_2$ is chosen so that the sequence contains repeated entries (for example $c_2 = a/N$ with integer $a$), distinct permutation orders can produce identical chirp matrices, shrinking the effective key space below $N!$ and invalidating the factorial-complexity argument.
The permutation could be reused as a lightweight session key: refreshing $k$ per transmission block changes the waveform completely without altering modulation order or coding, a property the paper does not develop.
Bit-error rate near 0.5 does not by itself prove zero leakage; a stronger guarantee would require a mutual-information or semantic-security analysis across all possible detectors, which the paper does not provide.

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

You should know this paper has a good core and a gap that kills the main claim. The new thing is applying the authors' chirp-permuted AFDM construction to physical-layer security, and the combinatorial analysis is the clean part: the derangement probabilities, the ℓ≤10 near-certainty, and the BER floor at 0.5 for a mismatched key are all correct and well presented. The effective channel argument (mismatched demodulation leaves a dense channel) is also right. The self-citation to [14] is not a problem; the waveform is genuinely theirs.

The problem is the threat model omits pilot symbols. Eve has perfect CSI and the only secret is the permutation order k. Coherent detection needs known reference symbols, and if those are public—standard in wireless—Eve can do this: invert her channel, apply Λ_c1 F_N, and divide by the known pilot symbols. That recovers e^{j2π c2 σ(j)^2} for each pilot j, which gives the permutation directly with linear operations. The exhaustive and quantum search analysis in Section IV becomes moot. If the authors intended pilots to be secret too, then k is not the only secret, and the N!-key argument no longer quantifies security. Either way the 'virtually perfect security' claim as written is unsupported.

Two smaller issues. The N! key count assumes every permutation gives a distinct chirp sequence; that fails for rational c2 like a/N because n^2 mod N has duplicates, so some keys collapse. And the query complexity for N=3300 is off: √(3300!) is about 10^5090, not 1.8×10^1654. Minor, but in a security paper the numbers should be right.

I would still send this to peer review. The flaw is not a non-starter; it's a missing piece of the model that a serious referee could catch and the authors could fix by either including a training-phase analysis or restricting the claim. The derangement analysis is publishable as part of a revised paper. For a reading group, it's actually a nice case study in why pilot secrecy matters.

Bottom line: engage with it, but don't take the security claims at face value. Send it to review with a request to check the threat model.

Referee Report

3 major / 4 minor

Summary. The paper proposes a physical-layer security scheme built on a chirp-permuted variant of AFDM, in which the permutation order of the second chirp sequence is a secret key shared by Alice and Bob. The authors claim that an eavesdropper with perfect CSI, and even co-located with Bob, cannot decode the data because the permutation key space has N! entries, making classical exhaustive search O(λN!), quantum Grover-style search O(√N!), and random guessing successful with probability at most 1/N!. The security argument is based on these complexity estimates and on a derangement analysis of near-guesses, supported by BER simulations showing that mismatched permutations yield BER around 0.5.

Significance. If the security claim were correct, the scheme would be an attractive low-cost physical-layer security mechanism for AFDM-based 6G systems, requiring no extra energy or hardware and offering claimed quantum resilience. The paper has some strengths: the AFDM effective-channel derivation is clear, the derangement formula in Eq. (14) is correctly applied, and the BER simulations illustrate the effect of mismatched permutations. However, the central security claim is undermined by a standard known-pilot attack that is not considered in the threat model, and the key-space size N! is not justified when chirp entries repeat. These issues are load-bearing rather than cosmetic, so the paper's main conclusion does not currently stand.

major comments (3)

[§III-C and §IV] The threat model assumes that the permutation key k is the only secret and that Eve has perfect CSI. Coherent reception requires known pilot or reference symbols, which the paper does not exclude or treat as secret. From Eq. (10b) and (8), Eve can compute s_hat = H_E^{-1} r_E ≈ A_k^{-1} x and then t = Λ_{c1} F_N s_hat ≈ Λ_{c2,k}^H x. For each known pilot position j, dividing t_j by the known pilot symbol x_j directly reveals the phase e^{j2π c2 (σ_k(j))^2}, from which the permutation σ_k is recovered by a linear operation followed by one division per pilot. This gives Eve the correct key without any search, contradicting the claimed 'virtually perfect' security in the Abstract and the exhaustive-search analysis of §IV-A.
[§IV-A and Eq. (6)] The paper states that there are N! possible permutation keys, but this is only true if the N entries of the chirp vector λ_{c2} are all distinct. For standard chirp parameter choices such as c2 = a/N with integer a, the phases e^{-j2π c2 n^2} can repeat with n (for example, N=4, a=1 gives 0,1,0,1), so many permutations produce the same diagonal matrix Λ_{c2,i} and hence the same waveform and the same key. Without a proof of distinctness, the factorial key-space count overestimates the effective security and the subsequent O(λN!) and O(√N!) complexity claims are not justified.
[§IV-A.2] The numerical quantum query complexity is incorrect. The paper reports O(√N!) ≈ O((2πN)^{1/4}(N/e)^{N/2}) ≈ O(1.8×10^1654) for N=3300, but direct Stirling evaluation gives about (2πN)^{1/4}(N/e)^{N/2} ≈ 10^5090, not 10^1654. Although the qualitative conclusion of infeasibility remains, the specific value quoted in the text and the implied quantitative argument are wrong and should be corrected.

minor comments (4)

[§IV-B.2 and Fig. 3] The figure legend text 'Eavesdropperwithℓ 5 10' is garbled; it should read 'Eavesdropper with ℓ ≤ 10'.
[Eq. (14)] The notation Pℓ is used with the subscript ℓ (ell), which is easily confused with the numeral 1 in print; consider using P_{\ell} consistently and defining the range ℓ ∈ {0,...,N} in the equation itself.
[Reference list] Reference [11] has an extraneous apostrophe and malformed punctuation in the author list; it should be cleaned up.
[§III-C] The symbol X is used both for the constellation set and for the vector set X^{N×1}; this dual use is confusing and should be distinguished, for instance by using a calligraphic symbol for the vector space.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the security analysis is self-contained and reduces to combinatorial search and guessing probabilities, not to fitted inputs or circular self-citations.

full rationale

The derivation chain is self-contained. The chirp-permuted AFDM waveform is defined directly in Eqs. (6)-(8), and the claim that permuting the second chirp sequence preserves the channel structure is proved in Eq. (9) via a similarity transformation of the intermediate effective channel, not imported from a cited result. The security analysis in Section IV reduces Eve's success to exhaustive search over the N! permutation keys (O(λN!) classical and O(√N!) quantum complexity) and to random-guess collision probabilities computed from derangements in Eq. (14); these are standard combinatorial facts applied to the model, not fitted parameters renamed as predictions. The BER simulations in Figure 3 illustrate the derived claims rather than constituting the derivation. The only self-citations, [14] for the chirp-permuted AFDM concept and [16] for the doubly-dispersive channel model, are contextual and non-load-bearing because the paper redefines and rederives the specific waveform properties it uses. A known-pilot attack that would recover the permutation with linear operations is a plausible modeling/completeness threat to the 'permutation as sole secret' claim, but that is a correctness concern, not a circular reduction, and therefore does not raise the circularity score.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

All entries rest on the system model in §II-III; the key-space and threat-model assumptions are load-bearing for the security claim.

assumptions (5)

ad hoc to paper The second chirp sequence Λ_{c2} has N distinct entries, so all N! permutations give distinct waveforms and keys.
Invoked implicitly in §III-A and §IV-A1 where the key space is set to N!. Not proven; for c2 = a/N with integer a, e^{-j2π c2 n^2} admits symmetries and duplicates, collapsing the key space.
domain assumption Eve's only viable strategies are exhaustive search and random guessing; she does not exploit known pilots or blind estimation.
The security analysis in §IV considers only these two strategies. Real coherent systems transmit known reference symbols which, with perfect CSI, allow Eve to infer the permutation directly.
domain assumption Both Bob and Eve have perfect CSI; Alice has none.
Stated in §II and used in §III-C for equalization and demodulation. Standard for PhySec analyses but favorable to the security claim.
standard math Standard combinatorial facts: derangement formula (14) and Stirling's approximation.
Used in §IV-A2 and §IV-B1 for search-space and guessing probabilities; these are correct.
standard math Grover/GAS can search the permutation space with O(√N!) queries given a valid oracle.
Cited from [17]-[19]; assumes Eve can construct an oracle that recognizes the correct permutation, which is non-trivial without known plaintext.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Chirp-Permuted AFDM for Quantum-Resilient Physical-Layer Secure Communications." pith.science (2026). https://pith.science/paper/6RMPHDTZ

@misc{pith2026250203289,
  author       = {Pith},
  title        = {Pith review of: Chirp-Permuted AFDM for Quantum-Resilient Physical-Layer Secure Communications},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6RMPHDTZ}},
  note         = {Machine review of arXiv:2502.03289}
}

read the original abstract

We present a novel physical-layer secure communications scheme based on a chirp-permuted variant of the affine frequency division multiplexing (AFDM) waveform recently proposed for sixth-generation (6G) systems, which ensures that eavesdroppers unaware of the correct chirp-permutation face significant challenges in signal detection, even with perfect channel state information (CSI) and co-location with the legitimate user. The security of the proposed scheme is studied in terms of the complexity required to find the correct permutation via classical and quantum search algorithms, analytically shown to be infeasible in both cases due to the factorially-scaling search space, and the probability of breach under the random-guess approach, also shown to be negligible.

Figures

Figures reproduced from arXiv: 2502.03289 by the authors.

**Figure 1.** Illustrations of the effective channels G∈CN×N with N = 64, of the AFDM [16], CP-AFDM (9), and mismatched CP-AFDM (11b), respectively in 2D (highlighting structure) and 3D (highlighting coefficients). The 2D axes denote the matrix indices of the effective channels G, and the vertical axis in 3D denotes the normalized magnitude of the non-zero channel coefficients. The above formulation reveals an important property … view at source ↗

**Figure 2.** Probabilities of Eve guessing a sequence which is ℓ-elements close to the true sequence, in terms of the PMF of the number of correct elements ℓ, and the CDF of the number of incorrect elements N − ℓ. The latter result is only a particular case of the more general problem of determining the likelihood that Eve makes a random guess close to the true permutation. To elaborate, let us suppose that Eve selects a permuta… view at source ↗

**Figure 3.** illustrates the BER performance obtained via numerical simulations, of the eavesdropper Eve and the legitimate user Bob, where the MMSE estimator for data detection is constructed using the random permutation key k ′ by Eve, and using the correct permutation key k by Bob. Two sets of numerical results are presented, for the remote eavesdropping (HE ̸= HB) and co-located eavesdropping (HE = HB) scenarios respectivel… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 19 canonical work pages

[14]

AFDM chirp-permutation-index modulation with quantum-accelerated codebook design,

H. S. Rou et al. , “AFDM chirp-permutation-index modulation with quantum-accelerated codebook design,” IEEE ASILOMAR Conf. , 2024

work page 2024
[1]

On the road to 6G: Visions, requirements, key technologies, and testbeds,

C. Wang et al. , “On the road to 6G: Visions, requirements, key technologies, and testbeds,” IEEE Commun. Surv. Tuts. , vol. 25, 2023

work page 2023
[2]

Integrated sensing and communications for 3D object imaging via bilinear inference,

H. S. Rou et al., “Integrated sensing and communications for 3D object imaging via bilinear inference,” IEEE Trans. Wireless Commun. , 2024

work page 2024
[3]

Security and privacy for 6G: A survey on prospective technologies and challenges,

V .-L. Nguyen et al., “Security and privacy for 6G: A survey on prospective technologies and challenges,” in IEEE Commun. Surv. Tuts., vol. 23, no. 4, pp. 2384-2428, 2021

work page 2021
[4]

Review of security methods based on classical cryptography and quantum cryptography,

S. Subramani and S. K. Svn, “Review of security methods based on classical cryptography and quantum cryptography,” Cybernetics and Systems, vol. 56, no. 3, pp. 302-320, 2025

work page 2025
[5]

Quantum speedup of the dispersion and codebook design problems,

K. Yukiyoshi et al., “Quantum speedup of the dispersion and codebook design problems,” IEEE Trans. Quant. Eng. , vol. 5, pp. 1–16, 2024

work page 2024
[6]

Post-quantum cryptography,

D. J. Bernstein and T. Lange, “Post-quantum cryptography,” Nature, vol. 549, no. 7671, pp. 188–194, 2017

2017
[7]

Post-Quantum Wireless-based Key Encapsulation Mechanism via CRYSTALS-Kyber for Resource- Constrained Devices,

M. A. Gonz ´alez de la Torre et al., “Post-Quantum Wireless-based Key Encapsulation Mechanism via CRYSTALS-Kyber for Resource- Constrained Devices,” IEEE Access, 2025

work page 2025

Show all 21 references

[8]

The Evolution of Quantum Key Distribution Networks,

Y . Cao et al., “The Evolution of Quantum Key Distribution Networks,” IEEE Commun. Surv. Tuts. , vol. 24, no. 2, pp. 839-894, 2022

2022
[9]

Physical-layer security in 6G networks,

L. Mucchi et al., “Physical-layer security in 6G networks,” IEEE Open. Journ. Commun. Soc. , vol. 2, pp. 1901–1914, 2021

1901
[10]

What physical layer security can do for 6G security,

M. Mitev et al., “What physical layer security can do for 6G security,” IEEE Open Journ. Veh. Tech. , vol. 4, pp. 375–388, 2023

2023
[11]

Artificial-noise-aided optimal beamforming in layered physical layer security,

W. Zhang et al., “Artificial-noise-aided optimal beamforming in layered physical layer security,” IEEE Commun. Lets. , vol. 23, no. 1, 2018

2018
[12]

Jamming attacks and anti-jamming strategies in wireless networks: A comprehensive survey,

H. Pirayesh et al., ’, “Jamming attacks and anti-jamming strategies in wireless networks: A comprehensive survey,”IEEE Commun. Surv. Tuts., vol. 24, no. 2, pp. 767–809, 2022

2022
[13]

Toward Chaotic Secure Communications: An RIS Enabled M-Ary Differential Chaos Shift Keying System With Block Interleaving,

X. Cai et al., “Toward Chaotic Secure Communications: An RIS Enabled M-Ary Differential Chaos Shift Keying System With Block Interleaving,” IEEE Trans. Commun. , vol. 71, no. 6, June 2023

2023
[15]

Affine frequency division multiplexing for next generation wireless communications,

A. Bemani et al. , “Affine frequency division multiplexing for next generation wireless communications,” IEEE Trans. Wireless Commun. , pp. 8214–8229, 2023

2023
[16]

H. S. Rou et al. , “From orthogonal time-frequency space to affine frequency-division multiplexing: A comparative study of next-generation waveforms for integrated sensing and communications in doubly disper- sive channels,” IEEE Sig. Proc. Mag. , vol. 41, no. 5, pp. 71–86, 2024

2024
[17]

Grover adaptive search for constrained polynomial binary optimization,

A. Gilliam et al. , “Grover adaptive search for constrained polynomial binary optimization,” Quantum, vol. 5, p. 428, 2021

2021
[18]

Grover adaptive search for maximum likelihood detection of generalized spatial modulation,

K. Yukiyoshi et al. , “Grover adaptive search for maximum likelihood detection of generalized spatial modulation,” in IEEE 100th Veh. Techno. Conf. Fall, 2024, pp. 1–5

2024
[19]

Accelerating grover adaptive search: Qubit and gate count reduction strategies with higher order formulations,

Y . Sano et al. , “Accelerating grover adaptive search: Qubit and gate count reduction strategies with higher order formulations,” IEEE Trans. Quan. Eng., vol. 5, pp. 1–12, 2024

2024
[20]

The quantum tortoise and the classical hare: A simple framework for understanding which problems quantum computing will accelerate (and which it will not),

S. Choi et al. , “The quantum tortoise and the classical hare: A simple framework for understanding which problems quantum computing will accelerate (and which it will not),” arXiv preprint 2310.15505 , 2023

2023 arXiv
[21]

Derangements and applications,

M. Hassani, “Derangements and applications,” Journ. Integer Seq. , vol. 6, no. 1.03.1, 2003

2003

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.

[1] [14]

AFDM chirp-permutation-index modulation with quantum-accelerated codebook design,

H. S. Rou et al. , “AFDM chirp-permutation-index modulation with quantum-accelerated codebook design,” IEEE ASILOMAR Conf. , 2024

work page 2024

[2] [1]

On the road to 6G: Visions, requirements, key technologies, and testbeds,

C. Wang et al. , “On the road to 6G: Visions, requirements, key technologies, and testbeds,” IEEE Commun. Surv. Tuts. , vol. 25, 2023

work page 2023

[3] [2]

Integrated sensing and communications for 3D object imaging via bilinear inference,

H. S. Rou et al., “Integrated sensing and communications for 3D object imaging via bilinear inference,” IEEE Trans. Wireless Commun. , 2024

work page 2024

[4] [3]

Security and privacy for 6G: A survey on prospective technologies and challenges,

V .-L. Nguyen et al., “Security and privacy for 6G: A survey on prospective technologies and challenges,” in IEEE Commun. Surv. Tuts., vol. 23, no. 4, pp. 2384-2428, 2021

work page 2021

[5] [4]

Review of security methods based on classical cryptography and quantum cryptography,

S. Subramani and S. K. Svn, “Review of security methods based on classical cryptography and quantum cryptography,” Cybernetics and Systems, vol. 56, no. 3, pp. 302-320, 2025

work page 2025

[6] [5]

Quantum speedup of the dispersion and codebook design problems,

K. Yukiyoshi et al., “Quantum speedup of the dispersion and codebook design problems,” IEEE Trans. Quant. Eng. , vol. 5, pp. 1–16, 2024

work page 2024

[7] [6]

Post-quantum cryptography,

D. J. Bernstein and T. Lange, “Post-quantum cryptography,” Nature, vol. 549, no. 7671, pp. 188–194, 2017

2017

[8] [7]

Post-Quantum Wireless-based Key Encapsulation Mechanism via CRYSTALS-Kyber for Resource- Constrained Devices,

M. A. Gonz ´alez de la Torre et al., “Post-Quantum Wireless-based Key Encapsulation Mechanism via CRYSTALS-Kyber for Resource- Constrained Devices,” IEEE Access, 2025

work page 2025