Digital signal processing at Bell Labs—Foundations for speech and acoustics research

Digital signal processing (DSP) is a fundamental tool for much of the research that has been carried out of Bell Labs in the areas of speech and acoustics research. The fundamental bases for DSP include the sampling theorem of Nyquist, the method for digitization of analog signals by Shannon et al., methods of spectral analysis by Tukey, the cepstrum by Bogert et al., and the FFT by Tukey (and Cooley of IBM). Essentially all of these early foundations of DSP came out of the Bell Labs Research Lab in the 1930s, 1940s, 1950s, and 1960s. This fundamental research was motivated by fundamental applications (mainly in the areas of speech, sonar, and acoustics) that led to novel design methods for digital filters (Kaiser, Golden, Rabiner, Schafer), spectrum analysis methods (Rabiner, Schafer, Allen, Crochiere), fast convolution methods based on the FFT (Helms, Bergland), and advanced digital systems used to implement telephony channel banks (Jackson, McDonald, Freeny, Tewksbury). This talk summarizes the key contributions to DSP made at Bell Labs, and illustrates how DSP was utilized in the areas of speech and acoustics research. It also shows the vast, worldwide impact of this DSP research on modern consumer electronics.