"Application of Correlation Analysis to the Detection of Periodic Signals in Noise" [Comment, with replies

Abstract

In the paper entitled "Application of correlation analysis to the detection of periodic signals in noise," by Y.W. Lee, T.P. Cheatham, Jr., and J.B. Wiesner (ibid., vol. 38, pp 1165-1171; Oct, 1950), several controvertible statements appeared. Because of the importance of this subject and the consequent possibility that the reader might easily arrive at erroneous conclusions, the commenting authors feel that these several statements should be amplified. In replying the original authors note the comments but fail to see that the statements concerned are incorrect or "controvertible." The commentors reply that they agree with the description of the special filter of North and Van Vleck and Middleton, but cannot follow the distinction which is made between this filter and a conventional filter in the penultimate paragraph of the reply. The original authors then note that these discussions primarily concern two statements of ours (as quoted in their reply) which Marchand, Leifer, and Holloway claim to be the "several controvertible statements" appearing in the paper. The validity of the first statement is not challenged after the reply. The original authors concludede, therefore, that it is not controvertible. With respect to the second statement, Marchand, Leifer, and Holloway do not disagree that the conventional filter described in the penultimate paragraph of our reply will theoretically achieve as much as a correlator in signal-to-noise ratio improvement. However, they do not follow the distinction between this filter and the North filter. The authors believe that the distinction between them has been made clear in their reply.