Mediaspace scheduled maintenance: Aug 25, 2026 07:00 - 12:00 AM. During this time, videos will be temporarily unavailable. Check status updates.

Lecture

Linear Estimation and Prediction

In course

EE-350: Signal processing

Dans ce cours, nous présentons les méthodes de base du traitement des signaux. ...

Description

This lecture covers the principles of linear estimation and prediction in signal processing, focusing on finding the signal in noisy signals, constructing linear filters, minimizing mean square error, and obtaining optimal filtering using Wiener filters. It also discusses the Wiener-Hopf equations, orthogonality, and the geometric interpretation of linear estimates. The lecture explores the Wiener filtering and prediction, the use of single pole filters, and the Wiener-Hopf equations for white noise. Additionally, it delves into the minimization of quadratic error, the Kalman filters for dynamical systems, and the importance of auto/inter-correlation functions in optimal linear prediction.

This video is available exclusively on Mediaspace for a restricted audience. Please log in to MediaSpace to access it if you have the necessary permissions.

Watch on Mediaspace

Instructors (2)

Jean-Philippe Thiran

Jean-Philippe Thiran was born in Namur, Belgium, in August 1970. He received the Electrical Engineering degree and the PhD degree from the Université catholique de Louvain (UCL), Louvain-la-Neuve, Belgium, in 1993 and 1997, respectively. From 1993 to 1997, he was the co-ordinator of the medical image analysis group of the Communications and Remote Sensing Laboratory at UCL, mainly working on medical image analysis. Dr Jean-Philippe Thiran joined the Signal Processing Institute (ITS) of the Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland, in February 1998 as a senior lecturer. He was promoted to Assistant Professor in 2004, to Associate Professor in 2011 and is now a Full Professor since 2020. He also holds a 20% position at the Department of Radiology of the University of Lausanne (UNIL) and of the Lausanne University Hospital (CHUV) as Associate Professor ad personam. Dr Thiran's current scientific interests include Computational medical imaging: acquisition, reconstruction and analysis of imaging data, with emphasis on regularized linear inverse problems (compressed sensing, convex optimization). Applications to medical imaging: diffusion MRI, ultrasound imaging, inverse planning in radiotherapy, etc.Computer vision & machine learning: image and video analysis, with application to facial expression recognition, eye tracking, lip reading, industrial inspection, medical image analysis, etc.

Pascal Frossard

Official source

Related concepts (32)

Kalman filter

For statistics and control theory, Kalman filtering, also known as linear quadratic estimation (LQE), is an algorithm that uses a series of measurements observed over time, including statistical noise and other inaccuracies, and produces estimates of unknown variables that tend to be more accurate than those based on a single measurement alone, by estimating a joint probability distribution over the variables for each timeframe. The filter is named after Rudolf E. Kálmán, who was one of the primary developers of its theory.

Wiener filter

In signal processing, the Wiener filter is a filter used to produce an estimate of a desired or target random process by linear time-invariant (LTI) filtering of an observed noisy process, assuming known stationary signal and noise spectra, and additive noise. The Wiener filter minimizes the mean square error between the estimated random process and the desired process. The goal of the Wiener filter is to compute a statistical estimate of an unknown signal using a related signal as an input and filtering that known signal to produce the estimate as an output.

Signal processing

Signal processing is an electrical engineering subfield that focuses on analyzing, modifying and synthesizing signals, such as sound, , potential fields, seismic signals, altimetry processing, and scientific measurements. Signal processing techniques are used to optimize transmissions, digital storage efficiency, correcting distorted signals, subjective video quality and to also detect or pinpoint components of interest in a measured signal. According to Alan V. Oppenheim and Ronald W.

Mean squared error

In statistics, the mean squared error (MSE) or mean squared deviation (MSD) of an estimator (of a procedure for estimating an unobserved quantity) measures the average of the squares of the errors—that is, the average squared difference between the estimated values and the actual value. MSE is a risk function, corresponding to the expected value of the squared error loss. The fact that MSE is almost always strictly positive (and not zero) is because of randomness or because the estimator does not account for information that could produce a more accurate estimate.

Filter (signal processing)

In signal processing, a filter is a device or process that removes some unwanted components or features from a signal. Filtering is a class of signal processing, the defining feature of filters being the complete or partial suppression of some aspect of the signal. Most often, this means removing some frequencies or frequency bands. However, filters do not exclusively act in the frequency domain; especially in the field of many other targets for filtering exist.

Related lectures (36)

Signal Processing: Noise Filtering and Signal Estimation

Explores noise filtering, signal estimation, and signal-to-noise ratio optimization through Wiener-Khintchine theorem and power spectral density.

Kalman Filter: Time Series

Covers structural modeling, state space models, and the Kalman filter in time series analysis.

Neural Signals and Signal Processing

Explores neural signal processing for brain-computer interfaces, including decoding techniques like Kalman filters and spike sorting.

Detection & Estimation

Covers the fundamentals of detection and estimation theory, focusing on mean-squared error and hypothesis testing.