Benoît Champagne

Department of Electrical & Computer Engineering

                            

Home Biography Research interests Publications Graduate students Courses taught Projects Demos

Research Interests

My research interests focus on the development and evaluation of new algorithms for the digital processing of information bearing signals, including speech, audio, data communication and radar/sonar signals. Known as Statistical Signal Processing, this field of study aims to devise and analyze the performance of computational procedures that can be used in practical engineering systems to draw inference (i.e. extract desired information) from imperfect or incomplete measurements. It makes intensive use of modern techniques from signal processing, applied probability and statistics, mathematical optimization, etc.

In my work, special emphasis is given to the study of efficient processing structures and algorithms for filtering, parameter estimation and tracking, as well as detection of signals in noise and interference (e.g.: echo, multi-path, reverberation). One of the driving goals of my research is to optimize the statistical performance of the new algorithms, while reducing their computational complexity so that they are amenable to real-time implementation. Targeted applications comprise multi-media communications, wireless communications and wireline (xDSL) transmissions. Research topics of current interest are listed below:

  • DSP fundamentals: filter design, digital filter banks, multi-rate systems, quantization effects, real-time DSP implementations;

  • Adaptive filtering: efficient filter structures (subband, frequency domain, etc), fast adaptation algorithms, performance analysis, equalization and echo cancellation;

  • Array sensing of communication signals: adaptive/robust beamforming, source location, algorithms, subspace tracking, collaborative processing, MIMO and relay systems;

  • Detection and parameter estimation: channel sensing and estimation, optimal signal detection, spatial and spectral frequency analysis, time delay estimation;

  • Speech and audio processing: audio classification, speech enhancement, acoustic echo cancellation, microphone arrays, talker location, voice activity detection.


[TSP laboratory]     [ECE Department]     [McGill University]

Last modified: 2010-06-16

Maintained by B. Champagne