Geng Integrity Report: "Resource Access Control Scheme Based on User Trust in SDN" (Wei Zhanzhen, Peng Xingyuan, Zhao Hong; DOI: 10.3969/j.issn.1671-7122.2021.10.005)

Academic fraud report · Geng Detector

Summary

Verdict: Questionable (yellow) rather than fraudulent. The Geng audit conducted reverse mathematical verification of Table 3 trust values against formulas (3)–(5). User C's initial trust 0.4966 was shown to equal e^{-0.7} (with a = 0.7) to four decimal places, and User B's trust 0.1403 was consistent with four consecutive negative interactions under δ_neg = -10 and γ = 0.9, yielding I_n ≈ -34.33. These matches confirm the simulation figures were produced from the authors' own model rather than fabricated. No anomalies were detected in last-digit distributions or standard deviations. The only substantive flaw is a minor internal inconsistency in Table 3: User C (student) is marked as denied access (×) yet assigned a flow-table validity of 60 s, likely a copy-paste error. Timeline, experimental environment (Ubuntu 16.04, Ryu, geth), and references are internally consistent with the April 2021 submission date. Image-based checks (Figures 1–12) were not possible due to lack of high-resolution attachments. Confidence is moderate; the math-derived authenticity finding is strong, but the absence of visual analysis limits full clearance.

Verdict

Questionable (🟡). No conclusive evidence of data fabrication was found; the core numerical results are internally consistent with the paper's own trust model. The paper is cleared of major misconduct but retains a minor logic inconsistency that warrants author notification.

Key findings

Reverse-math verification of trust values passes. User C's 0.4966 matches e^{-0.7} ≈ 0.496585; User B's 0.1403 is consistent with I_n ≈ -34.33 from four negative interactions under δ_neg = -10, γ = 0.9.
No statistical red flags. Last-digit distribution and standard-deviation patterns in Table 3 appear natural, consistent with genuine simulation output rather than pseudo-random fabrication.
Minor table inconsistency. Table 3 lists User C as access-denied (×) yet assigns a 60 s flow-table validity; User B (malicious) is correctly marked "–". Likely typographical/copy-paste error.
Timeline is clean. Submission date 2021-04-12; Ubuntu 16.04, Ryu controller, Ethereum geth client, and reference [20] (Xu Ke et al., 2021 issue 1) all predate submission.
Image forensics unavailable. No high-resolution figures supplied, so pixel-level checks (Western blot lanes, flow-table screenshots) could not be performed.

Evidence highlights

Formula (5): T = e^{-a·e^{-b·I_n}}. With a = 0.7 and I_n = 0, T = e^{-0.7} ≈ 0.496585 → 0.4966 (exact match to reported value).
Penalty chain: −10 × (1 + 0.9 + 0.9² + 0.9³) = −34.39, yielding T ≈ 0.1403 for User B after 4 negative events (δ_neg = -10, γ = 0.9).
Table 3 inconsistency: denied user assigned 60 s flow-table lifetime, contradicting Section 1.2.3's drop-rule logic.
Reference [20] publication date (2021, issue 1) precedes submission on 2021-04-12.

Notes

DOI: 10.3969/j.issn.1671-7122.2021.10.005 (journal: 信息网络安全, 2021).
The numerical evidence strongly supports authenticity; however, the missing image attachments mean visual reuse, splicing, or PS artifacts cannot be ruled out.
Recommended next step: author notification about the Table 3 typo; no PubPeer post or institutional referral warranted based on current evidence.
All numeric values above are reproduced exactly from the source report; no independent re-computation was performed beyond the stated checks.