arxiv: 2605.13331 · v1 · submitted 2026-05-13 · 💻 cs.IT · math.IT

Recognition: 2 theorem links

· Lean Theorem

Blind Recognition of Polar Codes Using Successive Cancellation List Decoding

Changwei Tu, Kai Niu, Xianzhao Feng, Yang Liu

Pith reviewed 2026-05-14 18:13 UTC · model grok-4.3

classification 💻 cs.IT math.IT

keywords blind recognitionpolar codessuccessive cancellation list decodinginformation setnon-cooperative communicationpath metric

0 comments

The pith

A list decoder identifies the unknown information set of a polar code by keeping the most reliable paths under competing frozen-bit and information-bit hypotheses at each position.

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

The paper introduces a blind recognition technique for polar codes whose length is known but whose information set is not. It runs successive cancellation list decoding while testing, at every bit, both the frozen-bit hypothesis and the information-bit hypothesis, then discards all but the L_list most reliable paths according to their average path metrics. The surviving path whose pattern best matches the observed statistical behavior of the source-side LLRs is declared the recognized information set. Because frozen bits produce decisions strongly biased toward zero while information bits produce nearly equiprobable decisions, the correct hypothesis set yields measurably better path metrics. Simulations for the (32,16), (64,32) and (128,64) polar codes show that the recognition success rate rises with list size and reaches at least a 2.5 dB improvement over earlier threshold-based methods when L_list equals 64.

Core claim

The central claim is that the distinct statistical behaviors of source-side decision LLRs—frozen bits favor zero while information bits are nearly equiprobable—can be exploited inside a successive cancellation list decoder: at each position the decoder expands paths under both hypotheses, retains only the L_list best paths by average path metric, and finally selects the information-set pattern belonging to the single most reliable surviving path.

What carries the argument

Successive cancellation list decoding performed under dual frozen-bit and information-bit hypotheses at each position, with path reliability measured by average path metric.

If this is right

Recognition success rate increases as the list size L_list grows.
For the (32,16), (64,32) and (128,64) polar codes the method yields at least 2.5 dB gain over prior threshold-based schemes when L_list = 64.
The scheme operates with known code length in non-cooperative settings without requiring channel bit-error-rate estimates.
Only the final most reliable path's information-set pattern is output as the recognition result.

Where Pith is reading between the lines

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

The same hypothesis-expansion idea could be tried with other soft-output decoders such as belief propagation if their path or message metrics also separate the two bit classes.
The method might extend to longer or punctured polar codes provided the statistical separation between frozen and information LLRs remains observable.
In a multi-user or multi-rate environment the technique could serve as a first stage to classify which polar code is being used before full decoding.

Load-bearing premise

The source-side decision LLRs exhibit distinctly different statistical behaviors for frozen bits (favoring zero) versus information bits (nearly equiprobable 0/1 decisions) under the channel conditions used in the simulations.

What would settle it

A Monte-Carlo test in which the average path metric of the correct information-set hypothesis is no better than that of incorrect hypotheses, or in which recognition success rate fails to increase when the list size is enlarged.

Figures

Figures reproduced from arXiv: 2605.13331 by Changwei Tu, Kai Niu, Xianzhao Feng, Yang Liu.

**Figure 3.** Figure 3: Performance comparison with P(128, 64) M = 500, it still provides about 2.5 dB gain at Pl = 0.5. V. CONCLUSION In this letter, a blind SCL-based recognition scheme for polar codes with known code length was proposed. By introducing frozen-bit and information-bit hypotheses as competing decoding paths, the proposed scheme establishes an effective connection between polar decoding and blind recognition, whi… view at source ↗

**Figure 2.** Figure 2: Performance comparison with P(64, 32) is M = 100, it achieves about 5 dB gain at Pl = 0.6. When M = 500, it still provides about 4.0 dB gain at Pl = 0.8 [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

read the original abstract

Blind recognition of polar codes remains challenging in non-cooperative scenarios, particularly for information-set recognition with known code length. Existing methods mainly rely on threshold decisions determined by the generator-matrix structure and channel bit error probability, without fully exploiting the soft information in received signals. In this letter, we propose a blind recognition method using successive cancellation list (SCL) decoding for polar codes with known code length. The proposed method exploits the distinct statistical behaviors of frozen and information bits in source-side decision log-likelihood ratios (LLRs) over multiple received vectors: frozen bits tend to favor zero decisions, whereas information bits exhibit nearly equiprobable $0/1$ decisions. Based on this property, the decoder expands candidate paths under the frozen-bit and information-bit hypotheses at each bit position, evaluates their reliabilities using the corresponding average path metrics, and retains only the $L_{\mathrm{list}}$ most reliable paths for subsequent recognition. Finally, the information-set pattern corresponding to the most reliable surviving path is selected as the recognition result. Simulation results show that the proposed scheme improves the recognition success rate as the list size increases. For the $(32,16)$, $(64,32)$, and $(128,64)$ polar codes, it achieves at least $2.5$ dB gain over the previous method when $L_{\mathrm{list}}=64$.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

SCL path expansion under frozen/info hypotheses gives a practical lift for blind polar recognition on short codes, but the claimed low-SNR gains rest on an LLR separation that noise should weaken.

read the letter

The new piece is straightforward: instead of fixed thresholds on the generator matrix, they run SCL while forcing each bit position to be either frozen or information, keep the L_list best paths by average metric, and declare the surviving pattern the recognized information set. That step is not in the earlier threshold papers they cite, and the simulations show the success rate rising cleanly with list size on the (32,16), (64,32), and (128,64) codes they tested. The gain reaches at least 2.5 dB at L_list=64, which is a usable improvement for a narrow receiver task. The method stays close to the existing decoder, so implementation cost looks low once you already have SCL code. The main weakness is the reliance on source-side LLRs behaving differently for frozen bits (strong zero bias) versus information bits (near 50/50). At the low-SNR region where the 2.5 dB figure appears, channel noise randomizes signs even for frozen bits, so the metric separation that drives the selection should shrink. The abstract gives no error bars, no exact channel model parameters, and no Monte-Carlo count, which leaves the central performance claim only partly supported. This is worth referee time for anyone working on non-cooperative polar-code interception or similar short-block recognition problems. It is not a theoretical advance, but the algorithmic tweak is concrete enough that a careful check of the low-SNR regime and the simulation details would be useful. I would send it out rather than desk-reject.

Referee Report

2 major / 2 minor

Summary. The paper proposes a blind recognition method for polar codes of known length that uses successive cancellation list (SCL) decoding. It exploits the statistical distinction in source-side decision LLRs across multiple vectors—frozen bits favor zero decisions while information bits are nearly equiprobable—and expands candidate paths under each hypothesis at every bit position, retaining the L_list most reliable paths according to average path metrics before selecting the information-set pattern of the best surviving path. Simulations for the (32,16), (64,32), and (128,64) polar codes report that recognition success rate increases with list size and that the scheme achieves at least 2.5 dB gain over a prior threshold-based method when L_list=64.

Significance. If the performance gains prove reproducible, the work would provide a concrete advance in non-cooperative polar-code recognition by replacing explicit threshold decisions with a soft-information-driven hypothesis search. The SCL framework naturally enumerates information-set hypotheses without requiring additional fitted parameters, which is a methodological strength.

major comments (2)

[Simulation results] Simulation results (abstract and §4): the claimed 2.5 dB gain for the three code lengths at L_list=64 is reported without error bars, the exact number of Monte-Carlo trials, or a precise statement of the channel model (e.g., AWGN variance or SNR definition). Because the central performance claim rests on these curves, the absence of these details leaves the magnitude and statistical significance of the improvement only partially supported.
[Proposed method] Proposed method (§3): the recognition procedure is justified by the assumption that source-side decision LLRs exhibit a clear zero bias for frozen bits versus equiprobable decisions for information bits. At the low-SNR operating points where the 2.5 dB gain is reported, channel noise can randomize LLR signs even for frozen bits, eroding the separation that drives path-metric differentiation. No histogram, table, or quantitative check of this separation is supplied for the relevant SNR range, making the load-bearing assumption unverified.

minor comments (2)

[Abstract] Abstract: the statement of simulation parameters is incomplete; adding the SNR range, trial count, and channel model would improve reproducibility.
[Notation] Notation: L_list is written as L_{list} in the text but L_{mathrm{list}} in the abstract; consistent typesetting throughout would aid readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address each major comment below and will revise the manuscript to strengthen the presentation of the simulation setup and the justification of the core assumption.

read point-by-point responses

Referee: [Simulation results] Simulation results (abstract and §4): the claimed 2.5 dB gain for the three code lengths at L_list=64 is reported without error bars, the exact number of Monte-Carlo trials, or a precise statement of the channel model (e.g., AWGN variance or SNR definition). Because the central performance claim rests on these curves, the absence of these details leaves the magnitude and statistical significance of the improvement only partially supported.

Authors: We agree that these details are required for full reproducibility and statistical assessment. In the revised manuscript we will state that all simulations use an AWGN channel with SNR defined as Eb/N0, employ 10,000 independent Monte-Carlo trials per SNR point, and display error bars corresponding to the standard error of the recognition success rate. These additions will appear in Section 4 and, space permitting, in the abstract. revision: yes
Referee: [Proposed method] Proposed method (§3): the recognition procedure is justified by the assumption that source-side decision LLRs exhibit a clear zero bias for frozen bits versus equiprobable decisions for information bits. At the low-SNR operating points where the 2.5 dB gain is reported, channel noise can randomize LLR signs even for frozen bits, eroding the separation that drives path-metric differentiation. No histogram, table, or quantitative check of this separation is supplied for the relevant SNR range, making the load-bearing assumption unverified.

Authors: The zero bias for frozen bits follows directly from the polar-code construction (frozen bits are fixed to 0) while information bits carry equiprobable data; the SCL path metric aggregates this distinction over multiple received vectors. Nevertheless, we acknowledge that an explicit check at the low-SNR regime would strengthen the exposition. In the revision we will add a figure in §3 that plots the empirical distribution of source-side decision LLRs (and the resulting average path metrics) for frozen versus information bits at representative low-SNR values (0 dB and 2 dB), confirming that a usable separation persists. revision: yes

Circularity Check

0 steps flagged

No significant circularity; method is algorithmically defined from SCL

full rationale

The paper defines its blind recognition procedure directly from the standard successive cancellation list (SCL) decoding algorithm by hypothesizing frozen vs. information bits at each position, expanding paths, and selecting the pattern with the highest average path metric. This exploits an observed statistical property of source-side decision LLRs (frozen bits favoring zero, information bits near-equiprobable), which is an input assumption drawn from channel behavior rather than derived from the recognition output. No parameters are fitted to the target information-set pattern, no equations reduce to their own inputs by construction, and the central claim is validated on independent Monte-Carlo trials for fixed codes. No self-citation forms a load-bearing uniqueness theorem or ansatz. The derivation chain is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The method rests on the standard polar-code construction that frozen bits are known to the encoder and on the channel-induced statistical separation of LLR decisions; no new entities are postulated.

free parameters (1)

L_list
List size is a tunable decoder parameter that trades complexity for recognition performance.

axioms (1)

domain assumption Frozen bits tend to favor zero decisions while information bits exhibit nearly equiprobable 0/1 decisions in source-side LLRs
Invoked in the abstract as the distinguishing property exploited by the path-expansion step.

pith-pipeline@v0.9.0 · 5541 in / 1287 out tokens · 44970 ms · 2026-05-14T18:13:42.891721+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear
the average metric increment under the frozen-bit hypothesis at index j is defined as Λ(F,j) = (1/M) Σ ln(1 + e^{-L(u_m,j)})

Reference graph

Works this paper leans on

10 extracted references

[1]

Channel polarization: A method for constructing capacity- achieving codes for symmetric binary-input memoryless channels,

E. Arıkan, “Channel polarization: A method for constructing capacity- achieving codes for symmetric binary-input memoryless channels,”IEEE Trans. Inf. Theory, vol. 55, no. 7, pp. 3051–3073, Jul. 2009

2009
[2]

TSG RAN WG1 AH1 NR,

3rd Generation Partnership Project (3GPP), “TSG RAN WG1 AH1 NR,” Athens, Greece, Feb. 2017

2017
[3]

List decoding of polar codes,

I. Tal and A. Vardy, “List decoding of polar codes,”IEEE Trans. Inf. Theory, vol. 61, no. 5, pp. 2213–2226, May 2015

2015
[4]

CRC-aided decoding of polar codes,

K. Niu and K. Chen, “CRC-aided decoding of polar codes,”IEEE Commun. Lett., vol. 16, no. 10, pp. 1668–1671, Oct. 2012

2012
[5]

Blind recognition algorithm of polar code based on information matrix estimation,

J. Liu, T. Zhang, H. Bai, and S. Ye, “Blind recognition algorithm of polar code based on information matrix estimation,”Syst. Eng. Electron., vol. 44, no. 2, pp. 668–676, 2022

2022
[6]

Blind identification of polar codes based on estimation and derivation approaches,

C. Yi, B. Pang, L. He, B. Ma, Y . Li, and F. C. M. Lau, “Blind identification of polar codes based on estimation and derivation approaches,”IEEE Commun. Lett., vol. 27, no. 2, pp. 414–418, Feb. 2023

2023
[7]

Blind recognition of polar code information bits based on multi-threshold voting and partial orders,

P. Xu, J. Liu, A. Wang, C. Yi, and Q. Li, “Blind recognition of polar code information bits based on multi-threshold voting and partial orders,”IEEE Commun. Lett., vol. 30, pp. 887–891, 2026

2026
[8]

Recognition of non-drilled polar codes based on soft decision,

Z. Wu, Z. Zhong, L. Zhang, and B. Dan, “Recognition of non-drilled polar codes based on soft decision,”J. Commun., vol. 41, no. 12, pp. 60–71, Dec. 2020

2020
[9]

Non-punctured polar code parameter recognition algorithm based on soft decision,

Y . Wang, C. Wang, X. Wang, and Z. Huang, “Non-punctured polar code parameter recognition algorithm based on soft decision,”Syst. Eng. Electron., vol. 45, no. 10, pp. 3293–3301, Oct. 2023

2023
[10]

Efficient design and decoding of polar codes,

P. Trifonov, “Efficient design and decoding of polar codes,”IEEE Trans. Commun., vol. 60, no. 11, pp. 3221–3227, Nov. 2012

2012