The special case of linear least squares is presented as an extremely useful estimation approach, in cases when the signal model can be written as a linear combination of known basis functions, with unknown weighting parameters. The linear least squares problem can be written as a matrix vector formulation and solved to yield the so-called normal equations. The video considers an example of estimating the Fourier coefficients of a signal modelled as a linear combinations of trigonometric functions. Finally, a numerical example is shown. The linear algebra manipulations are shown throughout in order to help the viewer manipulate similar types of equations, although a full geometric interpretation is not considered here.
PGEE11164 Probability, Estimation Theory, and Random Signals Lectures -- School of Engineering, University of Edinburgh. Copyright James R. Hopgood and University of Edinburgh, Scotland, United Kingdom (UK). 2020.
Institute for Digital Communications, Alexander Graham Bell Building, The King's Buildings, Thomas Bayes Road, Edinburgh, EH9 3JL. UK.