Sie sind auf Seite 1von 2

ECE 254: Detection Theory

Instructor:

Bill Hodgkiss

Tel: 858-534-1798
FAX: 858-822-0665
E-Mail: whodgkiss@ucsd.edu
Campus Mail: 0701

Course Description:

Hypothesis testing, detection of signals in white and colored Gaussian noise; estimation of signal parameters;
maximum-likelihood detection; resolution of signals; detection and estimation of stochastic signals; applications to
radar, sonar, and communications.

Summary of Topics Discussed:

(1) Introduction – SKE, SKEP, SKEPII problems.


(2) Summary of various PDFs – Gaussian and Chi-squared.
(3) Statistical decision theory framework – NP and Bayesian approaches.
(4) Deterministic signals – Matched filter and generalized matched filter.
(5) Random signals – Gaussian signal in Gaussian noise, Rayleigh fading signal.
(6) Composite hypothesis testing (uncertain parameters) – Bayesian and GLRT detection receivers.
(7) Deterministic signals with unknown parameters – amplitude, phase, frequency, and arrival time.
(8) Unknown noise parameters.

Homework/Projects:

Approximately one computer-oriented homework assignment will be made per week. These can be worked on in groups
and should be turned in as soon as possible for feedback. A mid-term and an end-term project will be assigned. These
should represent individual effort (i.e. should be considered as take-home exams) and assistance should not be given nor
received from anyone other than the instructor.

Grades:

No exams will be given. Grades will be assigned based on the weekly homework assignments and the mid/end-term
projects. The homework assignments count 1/3 and the mid/end-term projects count 1/3 each.

Text:

S. Kay. Fundamentals of Statistical Signal Processing - Vol II. Detection Theory. Prentice-Hall, 1998. ISBN 0-13-
504135-X.
ECE 254 – Detection Theory – Fall 2014

1. Are you graduate student or UCSD Extension (Concurrent Enrollment) student?

2. How much time do you think should be allocated for each of the following areas of course material?

a. Introduction – SKE, SKEP, SKEPII problems.


b. Summary of various PDFs – Gaussian and Chi-squared (Ch. 2).
c. Statistical decision theory framework – NP and Bayesian approaches (Ch. 3).
d. Deterministic signals – Matched filter and generalized matched filter (Ch. 4).
e. Random signals – Gaussian signal in Gaussian noise, Rayleigh fading signal (Ch. 5).
f. Composite hypothesis testing (uncertain parameters) – Bayesian and GLRT detection receivers (Ch. 6).
g. Deterministic signals with unknown parameters – amplitude, phase, frequency, and arrival time (Ch. 7).
h. Unknown noise parameters (Ch. 9).
i. Non-Gaussian noise (Ch. 10).

3. Please comment on the worth of the following:

a. Class text (S. Kay. Statistical Signal Processing – Vol II. Detection Theory.)

b. Reference books in the library

c. Lectures

d. TED web site with posted lecture notes, videos of lectures, and additional reference material

e. TA help sessions

f. Weekly computer homework

g. Mid-term and end-term projects

4. Are you working on the computer homework assignments using a class computer account, work or research
group computer, or personal computer?

5. What software are you using (MATLAB, other) and have you had any particular problems using it?

6. Did you use the ACMS Virtual Computing Labs and was this approach to software access beneficial?

7. Have you made use of the video podcasts of the class lectures? If so, are they useful and why? How could
they be improved?

8. What would you suggest as a better approach to the course material that would make it more interesting and
applicable to your interests?

9. What do you consider the most important weakness (fault) of this course?

10. Other critical remarks about the instructor and instruction?