DAVID F. DELCHAMPS
Associate Professor,
Electrical and Computer Engineering
329 Rhodes Hall
Phone: (607) 255-6447
E-mail: dave@ece.cornell.edu
B.S.E. 1976
(Princeton); M.S. 1977, Ph.D. 1982 (Harvard)
David F. Delchamps is originally from Morristown, NJ. Since receiving the
doctoral degree, he has been on the faculty of the School of Electrical
Engineering and a member of the Center for Applied Mathematics at Cornell,
where he is currently Associate Professor. He has received several
department- and college-level teaching awards, and was the recipient of an
NSF Presidential Young Investigator Award in 1984. He is a member of the
IEEE, the American Mathematical Society, Phi Beta Kappa, and Tau Beta Pi,
and is also the author of several technical articles and the book State
Space and Input-Output Linear Systems (Springer-Verlag, 1988). He does
research in the area of control and systems theory, with special emphasis on
the theory of estimation and control of nonlinear systems.
Research Interests and
Activities:
The behavior of a large
class of processes in engineering and the physical sciences can be described
fairly accurately in terms of mathematical models that take the form of
linear and nonlinear dynamical systems. Our current research focuses on
applying new mathematical techniques to the solution of certain fundamental
problems that arise in the theory of such models. Of special interest to us
are problems involving feedback control systems that have complicated
dynamical properties such as chaotic state evolution or pseudorandom
asymptotics. Such systems arise in many important applications; we have
found numerous examples of chaotic and pseudorandom asymptotic behavior in
dynamic analog-to-digital converters such as delta-sigma modulators and in
practical implementations of feedback control systems with
continuous-variable dynamics using digital controllers. The roundoff error
that one incurs because of quantized measurements of real numbers and
finiteprecision arithmetic are to blame for the complicated dynamical
phenomena; we have modeled these phenomena using techniques from the ergodic
theory of dynarnical systems that have heretofore had minimal impact in
signal processing and control systems contexts. We are also pursuing some
new approaches to the "intelligent" design of feedback control systems that
are subject to finite-precision effects such as those described in the
foregoing.
Selected Publications:
Delchamps, D. F. 1988. State Space and Input-Output
Linear Systems. New York: Springer-Verlag, Inc.
Delchamps, D. F. 1989. Extracting state information
from a quantized output record. Systems and
control Letters 13: 365-372.
Delchamps, D. F. 1990. Stabilizing a linear system with quantized state
feedback. IEEE Transactions on Automatic Control AC-35: 916-924.
Delchamps, D. F. October, 1990. Exact asymptotic statistics for sigma-delta
quantization noise. Proceedings of the 1990 Allerton Conference on
Communications, Control and Computing,
Urbana, IL. 573-601.
Delchamps, D. F. March, 1991. Spectral analysis of sigma-delta quantization
noise. Proceedings of the 1991 Conference on Information Sciences and
Systems, Baltimore, MD. 167-172.