OCE 561

Analysis of Oceanographic Data

Instructor: Professor Gopu R. Potty

Office: 115 Middleton Building (Bay Campus)

Phone: 874-6591; Fax: 874-6837

email: potty@oce.uri.edu

Office Hours: Anytime by appointment

Course Materials

• Homework
• Project
• Lectures

Course Goal: To provide each student with the tools, techniques, and knowledge to analyze random oceanographic data such as temperature and salinity time and space series, acoustic signals, currents, and other forms of data.

Disabilities: If you have a documented disability which may require individual accommodations, please make an appointment with me as soon as possible. We will discuss how to meet your individual needs to insure your full participation and fair assessment procedures.

Grading: This course will graded A-F based on 10 homework assignments (10%), 4 Matlab data analysis projects (30%), mid-term exam (25%), and final exam (35%).

Catalog description: Design of oceanic experiments to determine spatial and temporal sampling rate, precision, accuracy, signal-to-noise ratio, etc. Description of typical ocean data collection and analysis systems. Development of relevant techs. (Lec. 3)

Prerequisites: IME 411, MTH 451, or equivalent.

Textbooks: (1) Bendat and Piersol, Random Data. Analysis and Measurement Procedures, 3rd ed., Wiley, 2000.

(2) Hsu, Schaum’s outline of theory and problems in signals and systems, Schaum’s Outline Series, McGraw-Hill, 1995.

Course objectives: The student successfully completing this course will

• have a working understanding of signals and systems in both continuous and discrete time or space;

• have the capability to analytically compute Fourier series or transforms of simple signals in either discrete or continuous time/space, including the use and understanding the Matlab fft and fft2 programs;

• be able to design discrete and continuous time/space filters using Matlab including lowpass, bandpass, bandstop, and high pass filters and to use some of these filters to process real data;

• know how to work with stationary random processes and how to exploit their properties;

• know how to calculate an autocorrelation and spectrum of a random process, the cross correlation and cross spectrum of two random processes, and the coherence function, and will also know how to calculate and use confidence intervals and error bars;

• be competent in data acquisition procedures from instruments through analog filtering, through sampling, through discrete filtering to multichannel correlations and spectra;

• have a working understanding of time-frequency analysis techniques including short-time Fourier transforms, wavelet transforms, and other modern time-frequency transforms.

Topics covered

Week 1-2 Continuous and discrete time signals; linear systems, 1-D and 2-D signals.

Week 3-5 The many flavors of 1-D and 2-D Fourier transforms (CTFT, DFT, FFT, ZT, etc.)

Week 6-7 1-D and 2-D filter design and implementation.

Week 8 Probability and statistical principles.

Week 9-10 Stationary random processes, correlation and spectra.

Week 11 Statistical errors in estimates, confidence intervals.

Week 12-13 Data acquisition and processing; sampling, dynamic range.

Week 14-15 Time-frequency analysis, spectrograms, wavelets and other advanced topics.