In this course we learn to reduce data from two different space technologies, namely remote sensing satellites and radio interferometers. We will study a subset of the available algorithms that can be used to extract usefull information from the data collected by the aforementioned intruments. The course can be summed up as: data science in action. The course will be especially useful for students who wish to work at institutions like SANSA (South African National Space Sciences) or the SARAO (South African Radio Astronomy Observatory).
As examples of what is to come:
We will build a classifier that is able to classify MODIS (Moderate Resolution Imaging Spectroradiometer) pixels as either vegetation or settlement pixels.
We will build a small interferometric reduction pipeline whose final output will be an image of the Galactic center.
First Semester 2021
Just Show Up! Also taking CS 315 is recommended, but it is not a strict requirement.
After completing this course, students should be able to:
Understand the basics of error correction coding as it pertains to satellite communication (uplink and downlink).
Understand the basic physics that underpins remote sesning (the science of converting data of the earth recorded by distant platforms into usable information).
Understand the basics of sequential detection theory.
Understand the basics of classification (machine learning).
Understanding the basics of Fourier theory and how it pertains to interferometry.
Understand the basics of imaging, deconvolution and calibration as it pertains to interferometry.
Understand the basics of CASA (Common Astronomy Software Applications).
All lecture material (including journal papers, thesis chapters, slides etc...) will be made available on sunlearn.
There will be one lecture and one tutorial each week. Specific times are to be anounced.
Week 1: Belief Propogation
Week 2: No Lecture
Week 3: Remote Sensing
Week 4: Sequential Detection
Week 5: Machine Learning
Week 6: Fourier Theory
Week 7: Test 1
Week 8: Positional Astronomy
Week 9: Visibility Space
Week 10: Imaging
Week 11: Calibration
Week 12: CASA
Week 13: Test 2
This module is assessed by means of continuous assessment. The final mark for the module is calculated as:
Tutorials and Presentations: 25%
Written test A: 25%
Written test B: 25%
The instructor reserves the right to adjust marks and weights when it is deemed appropriate.
During tutorial classes students present (as groups – depending on enrolment numbers) the solutions to homework questions. The presentations will be graded for marks. Students will also be requiured to write a small tutorial test which will also count toward their tutorial mark.
Test: 60% (or 100% if class is too large for presentations)
During the course, students must complete two programming assignments (which they can complete in groups – depending on enrolment numbers). The assignments will be made available on sunlearn.
There will be two 3-hour, sit-down, closed-book written tests during the course of the semester. The test contents will be discussed in class.
The university's policy on plagiarism and collaboration applies throughout the course and is strictly enforced. It is the students' responsibility to familiarise themselves with the regulations in this regard. Students will also be asked to sign an integrity form for each assessment. Even perfectly correct assessments receive a mark of zero if the form is not signed.
Stellenbosch University / Faculty of Science / Mathematical Sciences / Computer Science