Signals & Noise

 
 

Research in physics and other fields often deals with signals which are buried in random and systematic noise.  This course covers techniques of experiment design, measurement, and analysis designed to avoid systematic error and optimize signal-to-noise ratio.   Extracting signals from noise, optimal filters, and examples of low-level detection spanning a range of subjects from laboratory physics, remote sensing, and astronomy will be discussed.  Many examples will come from detection of radiation (UV to sub-millimeter) and imaging, including inverse problems and data analysis. Theorists as well as experimentalists in several fields should find this course fun and useful. Students undertake a project of their own choosing utilizing any of these tools.

Syllabus

1. Noise sources, spectral analysis, fluctuation-dissipation

  1. 2.Experiment design, null experiments

  2. 3.Pathological science

  3. 4.Uncovering systematics

  4. 5.Noise reduction, isolation

  5. 6.Low level signal case studies

  6. 7.When noise is the signal

  7. 8.Image processing, inverse problems

  8. 9.Completeness vs efficiency

  9. 10. Optimal filtering, estimation, robustness

  10. 11. Sample bias, models, Monte Carlo simulation

Description

Coordinates

INSTRUCTOR  Tony Tyson

COURSE  PHY 123/253

OFFERED  Alternate years

WEB SITE