November 1, 2000 Editor, Journal of BUAA, Beijing 100083, People's Republic of China Dear Sir / Madam, I come across a paper published in the May 1998 issue of your journal; and it looks strikingly similar to another paper published four years earlier at an IEEE journal by other authors. I would like to bring the following information to your attention, for possible plagiarism by the authors of the 5/98 JBUAA paper. Thank you. With Regards, Kainam Thomas WONG, Ph.D. The Two Manuscripts Concerned: [Zol 94] Michael D. Zoltowski and Cherian P. Mathews, ¡°Real-Time Frequency and 2-D Angle Estimation Method with Sub-Nyquist Spatio-Temporal Sampling,¡± IEEE Trans. Signal Processing, vol. 42, no. 10, pp. 2781-2794, Oct. 1994. [Si 98] SI Deyi, LE Qiang, SHEN Shituan and LI Jingwen, ¡°Real-Time Frequency and 2-D Angle Estimation Method with Sub-Nyquist Sampling,¡± Journal of Beijing University of Aeronautics and Astronautics, pp. May, 1998. (submission date: July 17, 1997) Factual Comparison Between [Si 98] and [Zol 94]: (1) The abstract of [Si 98] starts off with ¡°A method is proposed ...¡±. This method, as the following demonstrates, is a summary of the method in [Zol 94]. (1a) The algorithmic development in section 1 of [Si 98] is a summary of the algorithmic development in sections II and III of [Zol 94]. (1b) The particular aliasing function (4) used in [Si 98] is identical as that in figures 3 and 4 of [Zol 94]. (1c) The section heading of section 1.1 in [Si 98] may be literally translated as ¡°frequency estimation¡¯s direct derivation¡±. The section heading of section 1.2 in [Si 98] may literally translated as ¡°frequency estimation¡¯s indirect derivation¡±. This terminology is analogous to ¡°direct ESPRIT approach¡± and ¡°indirect ESPRIT approach¡± in the lower right-hand-side column on p. 2785 of [Zol 94]. (2) [Zol 94] is cited in the reference section of [Si 98], along with two other references. The only place where [3] is cited in the body of [Si 98] is at the end of the first line in section 1.2. That line may be translated as ¡°First seek to solve for digital frequency f_i, then from the relationship between f_i and F_i seek to solve for F_i, can derive numerical frequency [3]¡±. (3) The title of [Si 98] is identical to [Zol 94] except omission of one word ¡°spatio- temporal¡±. A Possible Assessment: The abstract of [Si 98] claims the algorithm therein presented to be ¡°proposed¡± by the authors. This contrasts with the reality that this manuscript only recites [Zol 94] in summary form, along with the authors¡¯ own simple simulations. There is no acknowledgment or hint anywhere in [Si 98] that its ENTIRE algorithmic approach is from [Zol 94] --- not in its abstract, nor in its introduction, nor at the start of its algorithmic development, nor in its summary. Most readers may interpret the sole reference to [Zol 94] at the end of the first line in section 1.2 in [Si 98] to imply that (1) the algorithm in [Si 98] is indebted to [Zol 94] in only limited ways, and (2) the overall algorithmic scheme in [Si 98] is otherwise original from the authors. __________________________________________ Kainam Thomas WONG, Ph.D. Assistant Professor, Dept. of Electronic Engineering, Chinese University of Hong Kong, Shatin, NT, Hong Kong. Fax: (852) 2603-5558 (852 is Hong Kong's country code) ktwong@ee.cuhk.edu.hk http://www.ee.cuhk.edu.hk/~ktwong/