Levy教授来信澄清，陈军博士、洪伟教授等没有剽窃他
12-07
Levy教授来信澄清，陈军博士、洪伟教授等没有剽窃他的论文
2003年1月17日在新语丝网站上登出了一篇题为：“东南大学特聘教授洪伟偕弟子论文剽窃
开创历史之最”的匿名文章，认为1997年在IEEE MTT No.1上发表的署名陈军、洪伟等的论
文：“A new model of generalized inverter and its applications” 剽窃了著名教授
Ralph Levy 于1973年发表在同一刊物的论文：“A general design technique for prac
tical distributed reciprocal ladder networks”。其根据是：Levy教授1999年5月在I
EEE MTT上发表了他的评论信（comments，见附件二），而“两年”没有作者的答复函（A
uthor’s reply）。
事实是，1998年5月MTT编辑将Levy教授的评论信寄到了东南大学，但是通信作者陈军博
士已于1997年博士毕业后到了美国作博士后。当信件转到美国时，正好他从美国中部搬家
到西部。由于这两次耽搁，当他将作者答复信寄到MTT编辑部时，已过了编辑要求的截止日
期，结果导致1999年5月MTT发表Levy教授的评论信时没有将作者的答复信一起发表。之后
，陈军多次给编辑发emails，希望编辑能在以后期上将答复信发表出来，但没有结果。
平常阅读国际期刊的学者都知道，在刊物的后面经常发表评论信(comments)和作者的答
复信(Author’s reply)，这是非常正常的学术争论。
实际上，在Levy教授的评论信中并没有说剽窃他的论文，只是说在陈军、洪伟等的论文
中的一些公式经过三角变换后与他1973年论文中的一些公式是等价的，并认为作者可能不
知道他的文章。事实上，对于文中公式的等价性，只是K-Inverter部分的结果是等价的，
而N-Inverter部分的结果是Levy论文中所没有的。单就K-Inverter部分而言，陈军论文中
的结果也是通过不同的途径获得等价结果的。其实，只要仔细看一下这两篇论文的原文，
就可得出是否剽窃的结论。
昨天我们将关于两篇论文的看法用email直接发给了Levy教授本人，很快得到了他的回信
（见附件一）。尽管他对MTT论文评审制度有看法并对两篇论文创新点的本质差别还不完全
认同，但明确表示陈军等没有剽窃他的论文（Finally I am sure that you did not cop
y my 1973 paper），同时他表示：“遗憾的是我没有将广义变换器的概念另写为一篇论文
，而是隐含在1973年的论文之中，而这篇论文的题目又是不相关的，从而导致文献检索的
困难（It is unfortunate that I had not written the generalized inverter concep
t as a separate paper rather then embedding it in the 1973 paper with an unrel
ated title, making it difficult to trace in a literature search.）”。
因此，无论从Levy教授1999年的评论信，还是刚刚发来的电子邮件，都可以清楚地看到
这是一个学术争论问题，而非剽窃问题。
2003年1月23日
附件1：Levy教授给陈军博士的答复
Received: (from r.levy@juno.com) by m6.nyc.untd.com (jqueuemail) id
HN6MX8YN; Wed, 22 Jan 2003 17:07:28 EST
To: junchen@ieee.org
Date: Wed, 22 Jan 2003 14:09:11 -0800
Subject: Re: About your comments on 1999 MTT
From: "Ralph Levy" <r.levy@juno.com>
Dear Dr. Chen,
Thank you very much for your kind and considerate E-mail. One of the reasons m
y 1999 letter was submitted for publication was to rebuke the reviewers, who a
re the true culprits - they should have known better. However the standard of
reviewing for the Transactions has steadily worsened as the number of papers s
ubmitted for publication has increased. On the other hand you should be congra
tulated on having reproduced results which were original to your knowledge. It
is unfortunate that I had not written the generalized inverter concept as a s
eparate paper rather then embedding it in the 1973 paper with an unrelated tit
le, making it difficult to trace in a literature search.
I am not sure that I understand the essential difference between the symmetric
and asymmetric cases which you described. They are closely related since by c
ascading one by an ideal 90 degree phase shifter you obtain the other - see th
e first paragraph of the section "Generalized Impedance Inverter" on p. 520 of
the 1973 paper, and also Fig.2. Maybe I am missing something.
Finally I am sure that you did not copy my 1973 paper, especially since
you used different notation.
By the way I have consulted with National Semiconductor in El Segundo (
Los Angeles area). Best of luck in your career with them.
Sincerely,
Ralph Levy
附件2：Levy教授的评论信(IEEE Trans. MTT, Vol.47, No.5, p.655,1999)
Comments on “A New Model of Generalized Inverter and its Applications”
Ralph Levy
In the above paper 1, the generalized impedance inverter described with associ
ated theory is essentially identical to that published in 1973 [1]. Thus, the
inverter shown in Fig. 2 of the above paper1 is shown in [1, Fig. 3] and (9) a
nd (10) are identical to [1, eqs. (2)-(4)]. The correspondence is easily seen
by noting the difference in notation between the two papers, with a11, a12, a2
1, and a22 being identical to a, b, c, and d of [1]. Simple trigonometrical id
entities may be used to show that (9) of the above paper is identical to [1, e
q. (4)].
The example with the microstrip stepped junctions is similar to several demo
nstrations presented by this author and others in over 25 years since the publ
ication of [1], e.g., [1] contains three typical examples, and a paper in the
same issue of that TRANSACTIONS applies the generalized impedance inverter to
waveguide junctions [2]. More recently, the technique was used to design a cla
ss of miniature printed-circuit filters [3].
The Generalized impedance inverter is used to design a wide variety of unconve
ntional microwave filters and impedance transformers, which may include mixed
lumped and distributed elements and/or incommensurate transmission lines. A ty
pical (unpublished) example is the design of inhomogeneous waveguide transform
ers, where the step junctions are modeled as impedance inverters, whence an eq
uivalence between the transformer and an ideal homogeneous prototype may be ma
de which is exact at the upper frequency of the design band and proves to be v
ery good at lower frequencies. Minor changes to the intermediate waveguide wid
ths are carried out to give an optimum broad-band result. This technique is an
alternative to the traditional computer-aided design (CAD) method. Actually,
it may be used as a constraint in an optimization routine to ensure that all d
esigns generated in the program give a perfect result at a critical frequency,
resulting in a major reduction in optimization time.
In summary, the generalized impedance inverter is a fundamental concept in the
design of many complex ladder networks, enabling parasitic discontinuities an
d noncommensurate circuit elements to be taken into account automatically, rat
her than as an afterthought. Presumably the authors of the above paper1 were u
naware of [1].
REFERENCES
[1] R. Levy, “A generalized design technique for practical distributed recipr
ocal ladder networks,” IEEE Trans. Microwave Theory Tech., vol. MTT-21, pp. 5
19-526, Aug. 1973.
[2] R. Levy, “Tapered corrugated waveguide low-pass filters,” IEEE Trans. Mi
crowave Theory Tech., vol. MTT-21, pp. 526-532, Aug. 1973.
[3] J. C. Tipper and M. J. Shiau, “Circuit transformations for realization of
a class of miniature printed circuit filters,” in IEEE MTT-S Int. Microwave
Symp. Dig., May 1994, pp. 621-624.
_________________________
1 J. Chen, W. Hong, and C. H. Liang, IEEE Trans. Microwave Theory Tech., vol.
45, no. 1, pp. 132-135, Jan. 1997. 211.87.95.3
2003年1月17日在新语丝网站上登出了一篇题为：“东南大学特聘教授洪伟偕弟子论文剽窃
开创历史之最”的匿名文章，认为1997年在IEEE MTT No.1上发表的署名陈军、洪伟等的论
文：“A new model of generalized inverter and its applications” 剽窃了著名教授
Ralph Levy 于1973年发表在同一刊物的论文：“A general design technique for prac
tical distributed reciprocal ladder networks”。其根据是：Levy教授1999年5月在I
EEE MTT上发表了他的评论信（comments，见附件二），而“两年”没有作者的答复函（A
uthor’s reply）。
事实是，1998年5月MTT编辑将Levy教授的评论信寄到了东南大学，但是通信作者陈军博
士已于1997年博士毕业后到了美国作博士后。当信件转到美国时，正好他从美国中部搬家
到西部。由于这两次耽搁，当他将作者答复信寄到MTT编辑部时，已过了编辑要求的截止日
期，结果导致1999年5月MTT发表Levy教授的评论信时没有将作者的答复信一起发表。之后
，陈军多次给编辑发emails，希望编辑能在以后期上将答复信发表出来，但没有结果。
平常阅读国际期刊的学者都知道，在刊物的后面经常发表评论信(comments)和作者的答
复信(Author’s reply)，这是非常正常的学术争论。
实际上，在Levy教授的评论信中并没有说剽窃他的论文，只是说在陈军、洪伟等的论文
中的一些公式经过三角变换后与他1973年论文中的一些公式是等价的，并认为作者可能不
知道他的文章。事实上，对于文中公式的等价性，只是K-Inverter部分的结果是等价的，
而N-Inverter部分的结果是Levy论文中所没有的。单就K-Inverter部分而言，陈军论文中
的结果也是通过不同的途径获得等价结果的。其实，只要仔细看一下这两篇论文的原文，
就可得出是否剽窃的结论。
昨天我们将关于两篇论文的看法用email直接发给了Levy教授本人，很快得到了他的回信
（见附件一）。尽管他对MTT论文评审制度有看法并对两篇论文创新点的本质差别还不完全
认同，但明确表示陈军等没有剽窃他的论文（Finally I am sure that you did not cop
y my 1973 paper），同时他表示：“遗憾的是我没有将广义变换器的概念另写为一篇论文
，而是隐含在1973年的论文之中，而这篇论文的题目又是不相关的，从而导致文献检索的
困难（It is unfortunate that I had not written the generalized inverter concep
t as a separate paper rather then embedding it in the 1973 paper with an unrel
ated title, making it difficult to trace in a literature search.）”。
因此，无论从Levy教授1999年的评论信，还是刚刚发来的电子邮件，都可以清楚地看到
这是一个学术争论问题，而非剽窃问题。
2003年1月23日
附件1：Levy教授给陈军博士的答复
Received: (from r.levy@juno.com) by m6.nyc.untd.com (jqueuemail) id
HN6MX8YN; Wed, 22 Jan 2003 17:07:28 EST
To: junchen@ieee.org
Date: Wed, 22 Jan 2003 14:09:11 -0800
Subject: Re: About your comments on 1999 MTT
From: "Ralph Levy" <r.levy@juno.com>
Dear Dr. Chen,
Thank you very much for your kind and considerate E-mail. One of the reasons m
y 1999 letter was submitted for publication was to rebuke the reviewers, who a
re the true culprits - they should have known better. However the standard of
reviewing for the Transactions has steadily worsened as the number of papers s
ubmitted for publication has increased. On the other hand you should be congra
tulated on having reproduced results which were original to your knowledge. It
is unfortunate that I had not written the generalized inverter concept as a s
eparate paper rather then embedding it in the 1973 paper with an unrelated tit
le, making it difficult to trace in a literature search.
I am not sure that I understand the essential difference between the symmetric
and asymmetric cases which you described. They are closely related since by c
ascading one by an ideal 90 degree phase shifter you obtain the other - see th
e first paragraph of the section "Generalized Impedance Inverter" on p. 520 of
the 1973 paper, and also Fig.2. Maybe I am missing something.
Finally I am sure that you did not copy my 1973 paper, especially since
you used different notation.
By the way I have consulted with National Semiconductor in El Segundo (
Los Angeles area). Best of luck in your career with them.
Sincerely,
Ralph Levy
附件2：Levy教授的评论信(IEEE Trans. MTT, Vol.47, No.5, p.655,1999)
Comments on “A New Model of Generalized Inverter and its Applications”
Ralph Levy
In the above paper 1, the generalized impedance inverter described with associ
ated theory is essentially identical to that published in 1973 [1]. Thus, the
inverter shown in Fig. 2 of the above paper1 is shown in [1, Fig. 3] and (9) a
nd (10) are identical to [1, eqs. (2)-(4)]. The correspondence is easily seen
by noting the difference in notation between the two papers, with a11, a12, a2
1, and a22 being identical to a, b, c, and d of [1]. Simple trigonometrical id
entities may be used to show that (9) of the above paper is identical to [1, e
q. (4)].
The example with the microstrip stepped junctions is similar to several demo
nstrations presented by this author and others in over 25 years since the publ
ication of [1], e.g., [1] contains three typical examples, and a paper in the
same issue of that TRANSACTIONS applies the generalized impedance inverter to
waveguide junctions [2]. More recently, the technique was used to design a cla
ss of miniature printed-circuit filters [3].
The Generalized impedance inverter is used to design a wide variety of unconve
ntional microwave filters and impedance transformers, which may include mixed
lumped and distributed elements and/or incommensurate transmission lines. A ty
pical (unpublished) example is the design of inhomogeneous waveguide transform
ers, where the step junctions are modeled as impedance inverters, whence an eq
uivalence between the transformer and an ideal homogeneous prototype may be ma
de which is exact at the upper frequency of the design band and proves to be v
ery good at lower frequencies. Minor changes to the intermediate waveguide wid
ths are carried out to give an optimum broad-band result. This technique is an
alternative to the traditional computer-aided design (CAD) method. Actually,
it may be used as a constraint in an optimization routine to ensure that all d
esigns generated in the program give a perfect result at a critical frequency,
resulting in a major reduction in optimization time.
In summary, the generalized impedance inverter is a fundamental concept in the
design of many complex ladder networks, enabling parasitic discontinuities an
d noncommensurate circuit elements to be taken into account automatically, rat
her than as an afterthought. Presumably the authors of the above paper1 were u
naware of [1].
REFERENCES
[1] R. Levy, “A generalized design technique for practical distributed recipr
ocal ladder networks,” IEEE Trans. Microwave Theory Tech., vol. MTT-21, pp. 5
19-526, Aug. 1973.
[2] R. Levy, “Tapered corrugated waveguide low-pass filters,” IEEE Trans. Mi
crowave Theory Tech., vol. MTT-21, pp. 526-532, Aug. 1973.
[3] J. C. Tipper and M. J. Shiau, “Circuit transformations for realization of
a class of miniature printed circuit filters,” in IEEE MTT-S Int. Microwave
Symp. Dig., May 1994, pp. 621-624.
_________________________
1 J. Chen, W. Hong, and C. H. Liang, IEEE Trans. Microwave Theory Tech., vol.
45, no. 1, pp. 132-135, Jan. 1997. 211.87.95.3
其实老板们头衔极多，日理万机，不大可能有心思去细看自己博士写地文章，
估计是写文章的博士作者对以前的文献没完全看过，这也是有可能的。
无论如何此事对于俺们大家都是一个警示，
偶们写文章的时候一定得小心，须要尽可能地充分阅读文献才行啊。
166.111.146.216
相关文章：
- 教授是HFSS的第一代的CAD的设计者之一，不知道可有前途(05-08)
- 谢处方教授去世了(05-08)
- 国内有那些大学和教授是研究RFID的？(05-08)
- 国立台湾大学IEEE distinguished Lectuer王暉教授学术报告通知(05-08)
- 恭贺洪伟教授当选为IEEE fellow(05-08)
- 请问，清华大学有专门研究信号完整性分析的教授吗？(05-08)
射频专业培训教程推荐
