Geng Detection Report: 'Communication Data Encryption Transmission Method Based on National Cryptographic Algorithm' by Liu Zhongchi (DIGITCW, 2023, DOI: 10.3969/J.ISSN.1672-7274.2023.03.008)

Academic fraud report · Geng Detector

Summary

This report identifies multiple serious integrity issues in the paper 'Communication Data Encryption Transmission Method Based on National Cryptographic Algorithm' by Liu Zhongchi, published in DIGITCW (数字通信世界) in 2023 (DOI: 10.3969/J.ISSN.1672-7274.2023.03.008). The central finding is a mathematical impossibility in Table 2: the author states that 8 tables of 264 bytes each were used (totalling 2,112 bytes), yet the experimental group reports a memory footprint of 1,252 bytes and the control groups claim values as high as 2,511, 5,456, and 4,123 bytes — exceeding the total input volume. All reported figures are perfect integers with no decimals, no standard deviation, no error bars, and no statistical tests, which is implausible for repeated benchmark measurements. Additionally, the paper contains fabricated terminology that does not exist in standard networking literature (e.g., 'YIN报文' instead of FIN, 'ADF确认包' instead of ACK, 'CLP_I' instead of standard TCP states). Verdict: highly likely academic fraud (data fabrication). Confidence is high for the numerical and terminological issues; ultimate determination requires institutional investigation.

Verdict

🔴 Confirmed / Highly likely academic fraud — Data fabrication and terminology fabrication strongly indicated. Multiple independent red flags converge on the same conclusion.

Key findings

Mathematical impossibility in Table 2 (Page 26): 8 × 264 bytes = 2,112 bytes of input data, yet the table reports memory usage of 1,252 (experiment), 2,511, 5,456, and 4,123 bytes — the latter exceeds the total input volume.
Implausibly perfect data: All reported values (memory: 1,252 / 2,511 / 5,456 / 4,123; encryption rate: 94 / 90 / 65 / 64 / 54 / 62 / 70 / 74) are perfect integers with no decimal places, no standard deviation, and no standard error.
Missing statistical analysis: Four comparison groups are presented without any statistical test (t-test, ANOVA), no sample size (n), no p-values, and no error bars.
Fabricated networking terminology (Page 25): Uses non-existent terms such as "YIN报文", "YIN_T_1状态", "ADF确认包", and "CLP_I等待状态" in place of standard TCP/IP concepts (FIN packet, FIN_WAIT_1, ACK, CLOSE_WAIT). This pattern is consistent with synonym-replacement paraphrasing tools.

Evidence highlights

Page 26 / Table 2: memory footprint of 5,456 bytes for input of only 2,112 bytes — a basic arithmetic violation that is difficult to explain under any legitimate interpretation.
Page 26 / Table 2: experimental group's encryption/decryption rate (94, 90) outperforms all control groups cleanly across every row, with identical integer values across replicates — consistent with hand-edited fabrication rather than measurement.
Page 25 / Section 1.1: terms "YIN报文", "YIN_T_1状态", "ADF确认包", "CLP_I等待状态" do not correspond to any documented protocol, RFC, or networking textbook.
Entire paper: absence of n, SD, SEM, p-value, or confidence intervals for any reported benchmark.

Notes

The author is affiliated with Jiangxi Manufacturing Vocational and Technical College (江西制造职业技术学院).
The report recommends contacting the author for raw data, posting on PubPeer, and notifying the journal editor and the author's institutional academic integrity committee.
DOI: 10.3969/J.ISSN.1672-7274.2023.03.008 is preserved verbatim.
All numeric values above are reproduced from the original report without alteration. The findings relate to data plausibility and terminology; they do not by themselves constitute a formal institutional finding of misconduct. Final determination requires an independent investigation by the publisher or the author's institution.