Personal

Johan Sandberg

Research interests

Stochastic signal processing in general.
Audio analysis (speech and music) using Mel-like cepstrum (ordinary cepstrum, ordinary MFCC, and variants).
Statistical learning methods.
Estimation of spectrum and cepstrum.
Multitaper (Multiple Windows) methods.
Stochastic time-frequency analysis of non-stationary random processes. Especially discretization of time-frequency theory for processes in continuous time into theory for processes in discrete time.

Teaching

Spring 2010:

I will lecture the course Stationary Random Processes, FMS045/MAS210 (Stationära stokastiska processer), which will start the 15:th of March. Please contact me if you are interested in this course and have any questions.

Spring 2009:

Giving the course Stationary Random Processes, FMS045/MAS210 (Stationära stokastiska processer)

Publications

Journal Papers

J. Sandberg, M. Hansson-Sandsten, T. Kinnunen, R. Saeidi, P. Flandrin, and P. Borgnat “Multitaper Estimation of Frequency-Warped Cepstra with Application to Speaker Verification,” IEEE Signal Processing Letters, vol. 17, no. 4, pp. 343-346, 2010.
J. Sandberg and M. Hansson-Sandsten, “Optimal Stochastic Discrete Time-Frequency Analysis in the Ambiguity and Time-Lag Domain,” Signal Processing, vol. 90, no. 7, pp. 2203-2211, 2010.
J. Sandberg and M. Hansson-Sandsten, “Optimal Non-Parametric Covariance Function Estimation for any Family of Non-Stationary Random Processes,” submitted.
M. Hansson-Sandsten and J. Sandberg, “Mean Square Error Optimal Multiple Windows for Cepstrum Estimation,” submitted.
M. Hansson-Sandsten and J. Sandberg, “Optimization of weighting factors for multiple window spectrogram of event related potentials,”, accepted for publication in EURASIP Journal on Advances in Signal Processing.
J. Sandberg and M. Hansson-Sandsten, “A Comparison Between Different Discrete Ambiguity Domain Definitions in Stochastic Time-Frequency Analysis,” IEEE Transactions on Signal Processing, vol. 57, no. 3, pp. 868-877, 2009.

Conference Papers

T. Kinnunen, R. Saeidi, J. Sandberg, and M. Hansson-Sandsten, “What Else is New Than the Hamming Window? Robust MFCCs for Speaker Recognition via Multitapering,” submitted.
J. Sandberg and M. Sandsten, “Approximate Optimal Periodogram Smoothing for Cepstrum Estimation using a Penalty Term,” accepted for publication in EUSIPCO, European Signal Processing Conference 2010, Aalborg, Denmark. August 2010.
J. Sandberg and M. Sandsten, “Optimal Wigner Cross-Spectrum estimation,” EUSIPCO, European Signal Processing Conference 2009, Glasgow, UK. August 2009, pp. 2298-2302.
M. Sandsten and J. Sandberg, “Optimization of Weighting Factors for Multiple Window Time-Frequency Analysis,” EUSIPCO, European Signal Processing Conference 2009, Glasgow, UK. August 2009, pp. 2283-2287.
M. Sandsten and J. Sandberg, “Optimal Cepstrum Estimation Using Multiple Windows,” ICASSP, IEEE International Conference on Acoustics, Speech, and Signal Processing, Taipei, Taiwan. April 2009, pp. 3077-3080.
J. Sandberg and M. Hansson, “Coherence Estimation between EEG Signals using Multiple Window Time-Frequency Analysis compared to Gaussian Kernels,” EUSIPCO, European Signal Processing Conference, Florence, Italy. Sept. 2006. Download poster presentation: EUSIPCO06 poster (pdf: 38 mb)
J. Sandberg, M. Lindgren, and M. Hansson, “Detecting MMN in Infants EEG with Singular Value Decomposition,” EMBC, 27th Annual Int. Conf of the IEEE Engineering in Medicine and Biology Society, Shanghai, China. Sept. 2005, pp. 4227-4230.
M. Hansson and J. Sandberg, “Multiple Windows for Estimation of Locally Stationary Transients in the Electroencephalogram,” EMBC, 27th Annual Int. Conf of the IEEE Engineering in Medicine and Biology Society, Shanghai, China. Sept. 2005, pp. 7293-7296.

Theses

J. Sandberg, “Discrete Stochastic Time-Frequency Analysis and Cepstrum Estimation,” Ph.D. Thesis in Mathematical Statistics, 2010. ISBN: 978-91-628-8080-4. Advisor: Professor Maria Hansson-Sandsten, Opponent: Professor Magnus Jansson, Royal Institute of Technology, Stockholm. View presentation slides:
Ph.D. slides (pdf: 3 mb)
Ph.D. slides (flash)
J. Sandberg, “Ambiguity Domain Definitions and Covariance Function Estimation for Non-Stationary Random Processes in Discrete Time,” Licentiate Thesis in Mathematical Statistics, 2008. Advisor: Professor Maria Hansson-Sandsten, Opponent: Professor Martin Stridh.
J. Sandberg, “Detecting MMN in Infants EEG with Bootstrap and SVD,” Master Thesis in Mathematical Statistics, 2005