Signal Processing Analysis of the Kennedy Assassination Tapes
R.C. Agarwal, R. L. Garwin, and B. L. Lewis
IBM T. J. Watson Research Center
Yorktown Heights, NY 10598
Abstract: On the basis of analysis by two research groups of impulses identified as "shots" on a 5-minute continuous recording of Dallas Police Department Channel I, the House Select Committee on Assassinations concluded that "scientific acoustic evidence establishes a high probability that two gunmen fired at President John F. Kennedy." The Report of the Committee on Ballistic Acoustics of the National Academy of Sciences (May, 1982) demonstrated that the signals in question were recorded about a minute after the assassination, that they contained an image of words (from DPD Channel II) spoken well after the assassination, and that the overlying words were recorded from the radio together with the impulses (and hence were not added later). This report documents the objective computer signal-processing approach we used in correlating two-dimensional sound spectrograms, expands on the effect of Channel-I heterodynes on Channel-I audio and on the Channel-II imprint, and details the use of fortuitous Channel-II "brief tones" to measure the transmission from Channel II to Channel I. We demonstrate also the [sic] the notorious "bell" on Channel I is present on Channel II at the same instant, as determined by a cross-talk tiepoint between channels, but is not cross-talk. It must be electrical interference superimposed on both channels. Our digital data have been deposited with the National Academy of Sciences.
Back to IBM report table of contents
Back to IBM report table of contents
Back to archive of acoustical evidence
Back to JFK reports and documents menu