{"id":"2f4dbe73-b0c9-4e1d-9d5c-66f32acccf5f","arxiv_id":"2502.03289","paper_version":2,"verdict":"REJECT","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"A secret permutation of the second chirp sequence in AFDM is proposed as a physical-layer security key, with security argued from factorial search-space size and derangement-based guessing probabilities.","lead":"This paper proposes a wireless security scheme in which the transmitter and receiver share a secret permutation of a chirp sequence inside the AFDM waveform, creating a factorial key space the authors argue is too large for exhaustive search, even with quantum computers. Why read it: to see whether a simple waveform-level 'permutation key' can really provide quantum-resilient physical-layer security without extra energy or hardware, and where that argument fails.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Known pilot/reference symbols let Eve recover the permuted chirp phases with linear operations, so the 'permutation as sole secret' claim is unsupported.","rationale":"The reader's weakest_assumption identifies exactly the load-bearing gap that I find most damaging: the paper never analyzes an adversary with access to known reference/pilot symbols, even though the paper assumes perfect CSI and coherent demodulation, both of which normally require such reference symbols in practice. Under the paper's own statement that the chirp-permutation order is the only secret, the pilot/reference values are either known to Eve or they are additional secrets. If they are known, Eve can strip off the channel and the first chirp, expose the permuted second-chirp phases on pilot subcarriers, and recover the permutation directly; no factorial search is needed. If they are secret, then the claim that only the permutation is secret is false, and no security analysis is provided for the actual secret set. Either way the central conclusion in §IV is unsupported. The other concerns in the reader's verdict are real but secondary. The repeated-chirp-entry issue (e.g., c2 = a/N producing duplicate diagonal entries) would further reduce the effective key space, but even with an injective chirp the pilot attack succeeds, so it is not the single most load-bearing issue. The numerical error in the quantum query complexity, O(1.8×10^1654) versus the correct order near 10^5090 for N=3300, is an arithmetic mistake but does not change the qualitative conclusion that factorial search is infeasible; it is therefore not decisive. The absence of an information-theoretic secrecy analysis also matters, but the pilot-based recovery is more directly fatal to the paper's stated threat model. Because the load-bearing concern lands and the reader already reached REJECT, the verdict should remain unchanged.","tokens_in":9492,"tokens_out":5924,"duration_ms":60971,"concrete_test":"Simulate the exact system model with N=64, public c1 and c2 chosen so that n ↦ e^{j2π c2 n^2} is injective on {0,...,N−1}, a random secret permutation k, a full training block of known pilot symbols x, and Eve given perfect CSI and the same received signal as Bob. Compute t = Λ_{c1} F_N H^{-1} r and divide each element by the known pilot symbol; invert the resulting phases to recover σ_k. If this succeeds at high SNR, the exhaustive-search and random-guess analyses in §IV are not the relevant attack surface, and the claimed 'virtually perfect security' is not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (Abstract, §IV) is that with the secret chirp-permutation order k as the only secret, Eve with perfect CSI and co-location cannot decode. This fails once the reference/pilot symbols required for coherent reception are considered. Coherent AFDM reception needs known pilot/preamble symbols; if these are public, as is standard and as the paper appears to assume (only k is secret), Eve can compute s_hat = H_E^{-1} r_E ≈ A_k^{-1} x from (10b), then t = Λ_{c1} F_N s_hat ≈ Λ_{c2,k}^H x using (8). For every pilot position j, dividing t_j by the known pilot symbol x_j yields the diagonal phase e^{+j2π c2 (σ_k(j))^2}. With a full training block this gives the entire permutation σ_k directly, reducing the claimed O(λN!) classical or O(√N!) quantum search (Section IV-A) to a linear operation plus one division per pilot. The same argument applies in the co-located case, since it uses only perfect CSI and known pilots. The paper's security analysis (Eq. 14, Fig. 3) considers only random-guess and exhaustive-search adversaries; it never includes a reference-signal-equipped adversary. If the authors intended the pilots themselves to be secret, then the statement that the chirp-permutation order is the only secret is false, and the security analysis must instead quantify the residual uncertainty from partial pilot knowledge, which the paper does not do.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","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.","tokens_in":9688,"tokens_out":4117,"duration_ms":38807,"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":[{"comment":"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.","section":"§III-C and §IV"},{"comment":"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.","section":"§IV-A and Eq. (6)"},{"comment":"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.","section":"§IV-A.2"}],"minor_comments":[{"comment":"The figure legend text 'Eavesdropperwithℓ 5 10' is garbled; it should read 'Eavesdropper with ℓ ≤ 10'.","section":"§IV-B.2 and Fig. 3"},{"comment":"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.","section":"Eq. (14)"},{"comment":"Reference [11] has an extraneous apostrophe and malformed punctuation in the author list; it should be cleaned up.","section":"Reference list"},{"comment":"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.","section":"§III-C"}],"recommendation":"reject","confidential_remarks":"The known-pilot attack described in my major comment is, in my view, decisive for the paper's central claim. Because the authors explicitly state that the chirp permutation is the only secret, the presence of standard pilots in a coherent waveform immediately permits Eve to recover the permutation. This is a modeling gap that is not a small fix: either the pilots must be made secret (which changes the key model and the security analysis) or the security claims must be substantially weakened. The repeated-entry issue with the N! count further weakens the quantitative security analysis. Given these load-bearing problems, I do not see how a routine revision would preserve the paper's main contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","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.\n\nThe 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.\n\nTwo 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.\n\nI 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.\n\nBottom 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.","headline":"Clean combinatorial core, but the 'virtually perfect security' claim fails against a known-pilot attack that recovers the permutation directly.","tokens_in":10308,"tokens_out":3840,"would_cite":false,"duration_ms":37079,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"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.","keywords":["physical-layer security","AFDM","chirp-permutation","quantum resilience","wireless communications","beyond 5G","Grover adaptive search","derangements"],"falsifier":"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.","tokens_in":9184,"feed_emoji":"🔐","tokens_out":12563,"duration_ms":104916,"temperature":0.7,"pith_summary":"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.","feed_headline":"A secret chirp shuffle blocks even quantum-accelerated eavesdroppers","feed_subtitle":"The key space is N! permutations; a wrong guess leaves the eavesdropper with BER near 0.5.","key_machinery":"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.","core_discovery":"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.","pith_inferences":["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."],"forward_implications":["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."],"supporting_citations":[{"why":"Supplies the chirp-permuted AFDM concept and the codebook-domain index modulation that the security scheme converts into a secret-key mechanism.","marker":"[14]"},{"why":"Defines the AFDM waveform and its DAFT/IDAFT structure, including the property that the second chirp can be varied without losing delay-Doppler diversity.","marker":"[15]"},{"why":"Provides the doubly-dispersive circular-convolution channel model and the effective-channel representation used to compare matched and mismatched demodulation.","marker":"[16]"},{"why":"Gives the Grover adaptive search algorithm and its O(√N!) query complexity, the basis for the quantum eavesdropper analysis.","marker":"[17]"},{"why":"Shows Grover adaptive search applied to maximum-likelihood detection in wireless systems, the template for a quantum-accelerated eavesdropper.","marker":"[18]"},{"why":"Provides qubit and gate-count reduction strategies for Grover adaptive search, supporting the estimate of quantum resources needed for the permutation search.","marker":"[19]"},{"why":"Supplies the projected qubit counts available by 2040, used to argue that N=3300 CP-AFDM is quantum-resilient on that timescale.","marker":"[20]"},{"why":"Provides derangement counts and the formula P_ℓ = D_{N−ℓ}/N! used for the random-guess probability analysis.","marker":"[21]"},{"why":"Supplies the 5G/6G subcarrier-count context (up to N=3300) used to evaluate the factorial key space at practical system sizes.","marker":"[1]"}],"fun_headline_variants":["Quantum-proof chirp shuffle keeps secrets safe","N! permutations: a quantum search can't crack it","Chirp-permuted AFDM: eavesdroppers hit BER 0.5","Secret chirp order foils even quantum algorithms"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"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.","fun_headline_variants_meta":{"raw":{"variants":["Quantum-proof chirp shuffle keeps secrets safe","N! permutations: a quantum search can't crack it","Chirp-permuted AFDM: eavesdroppers hit BER 0.5","Secret chirp order foils even quantum algorithms"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000142,"raw_usage":{"total_tokens":1179,"prompt_tokens":967,"completion_tokens":212,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":583,"completion_tokens_details":{"reasoning_tokens":143}},"tokens_in":583,"tokens_out":212,"duration_ms":2893,"temperature":1.0,"reasoning_tokens":143,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T05:15:01.421601+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"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.","supporting_citations":[{"cited_title":"AFDM chirp-permutation-index modulation with quantum-accelerated codebook design,","cited_arxiv_id":null,"evidence_quote":"Supplies the chirp-permuted AFDM concept and the codebook-domain index modulation that the security scheme converts into a secret-key mechanism."},{"cited_title":"Affine frequency division multiplexing for next generation wireless communications,","cited_arxiv_id":null,"evidence_quote":"Defines the AFDM waveform and its DAFT/IDAFT structure, including the property that the second chirp can be varied without losing delay-Doppler diversity."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the doubly-dispersive circular-convolution channel model and the effective-channel representation used to compare matched and mismatched demodulation."},{"cited_title":"Grover adaptive search for constrained polynomial binary optimization,","cited_arxiv_id":null,"evidence_quote":"Gives the Grover adaptive search algorithm and its O(√N!) query complexity, the basis for the quantum eavesdropper analysis."},{"cited_title":"Grover adaptive search for maximum likelihood detection of generalized spatial modulation,","cited_arxiv_id":null,"evidence_quote":"Shows Grover adaptive search applied to maximum-likelihood detection in wireless systems, the template for a quantum-accelerated eavesdropper."},{"cited_title":"Accelerating grover adaptive search: Qubit and gate count reduction strategies with higher order formulations,","cited_arxiv_id":null,"evidence_quote":"Provides qubit and gate-count reduction strategies for Grover adaptive search, supporting the estimate of quantum resources needed for the permutation search."},{"cited_title":"Derangements and applications,","cited_arxiv_id":null,"evidence_quote":"Provides derangement counts and the formula P_ℓ = D_{N−ℓ}/N! used for the random-guess probability analysis."},{"cited_title":"On the road to 6G: Visions, requirements, key technologies, and testbeds,","cited_arxiv_id":null,"evidence_quote":"Supplies the 5G/6G subcarrier-count context (up to N=3300) used to evaluate the factorial key space at practical system sizes."}],"review_version":1}