Die Sprechzeiten für die IST Studienangelegenheiten: Dienstags 15:00 bis 16:30 Uhr
Veranstaltungen im Wintersemester
Veranstaltungen im Sommersemester
Han, S.; Pflug, F.; Fingscheidt,T.
in Proc. of 21st European Signal Processing Conference (EUSIPCO), Marrakech, Morocco, Sep. 2013
Mobile wideband speech communication (HD Voice) is more and more available in the past years, primarily in 3G networks. The specifics of mobile communication --- even if it is packet-switched --- is that received frames with residual bit errors after channel decoding must not necessarily be marked as lost, instead they may be marked as bad (bad frame indicator, BFI). In this work we present how to exploit the information of a soft input (i. e., a log-likelihood ratio input) within the Adaptive Multirate Wideband (AMR-WB) speech decoder, allowing a more robust error concealment as compared to the 3GPP Recommendation. Log-likelihood ratios may be taken from a soft-output channel decoder, or, as in our generic simulation, directly from the demodulator, without the need of a BFI. Since error concealment is non-mandatory, chipset manufacturers are free to implement this alternative speech decoding scheme still in a standard-compliant fashion.
Han,S.; Fingscheidt,T.
in Proc. of 22nd European Signal Processing Conference (EUSIPCO), Lisbon, Portugal, Sept. 2014
Lloyd-Max quantization (LMQ) is a widely used scalar nonuniform quantization approach targeting for the minimum mean squared error (MMSE). Once designed, the quantizer codebook is fixed over time and does not take advantage of possible correlations in the input signals. Exploiting correlation
in scalar quantization could be achieved by predictive quantization, however, for the price of a higher bit error sensitivity. In order to improve the Lloyd-Max quantizer performance for correlated processes without encoder-sided prediction, a novel scalar decoding approach utilizing the correlation of input signals is proposed in this paper. Based on previously received samples, the current sample can be predicted a priori. Thereafter, a quantization codebook adapted over time will be generated according to the prediction error probability density function. Compared to the standard LMQ, distinct improvement is achieved with our receiver in error-free and error-prone transmission conditions, both with hard-decision and soft-decision decoding.
Han,S.; Fingscheidt,T.
in Proc. of 10th International ITG Conference on Systems, Communications and Coding, Hamburg, Germany, Feb. 2015
Scalar non-uniform Lloyd-Max quantization (LMQ) is optimized for the minimum mean squared error satisfying the centroid condition, stating that the reconstruction level is the centroid of the area of the signal probability density function (PDF) in the respective quantization interval. For a given signal PDF, reconstruction levels are therefore time-invariant. Without predictive means at the encoder, scalar LMQ does not exploit source correlation at all. With the purpose of improving LMQ performance for correlated processes, we propose a novel scalar decoding approach employing a receiver-sided predictor inside a feedback loop. Based on the standard LMQ decision levels and utilizing the prediction error PDF shifted by the predictor output, a receiver-sided time-variant quantization codebook can be generated according to the centroid condition. Moreover, we give an analytical derivation of the variance of the prediction error. Simulations over an additive white Gaussian noise (AWGN) channel in error-free and error-prone transmission conditions show that the proposed approach outperforms the standard LMQ by about 1.0 dB in terms of signal-to-noise ratio (SNR), especially for low rate quantization and highly correlated source processes.
Han,S.; Fingscheidt,T.
in Proc. of 11th ITG Conference on Speech Communication, Erlangen, Germany, Sept. 2014
The well known scalar non-uniform Lloyd-Max quantizer (LMQ) is optimized for the minimum mean squared error (MMSE). Its quantization codebook reconstruction levels are determined by the centroid of the signal probability
density function (PDF) in a given quantization interval. Redundancy from correlated source processes is not used at all. To improve scalar quantizer performance for correlated processes, a novel decoding approach is proposed in this paper, employing a receiver-sided optimized predictor along with a standard LMQ encoder. Based on the prediction error PDF, the LMQ quantization codebook reconstruction levels are adapted over time. Moreover, a full search for the optimal values of the predictor coefficient and the variance of the prediction error is carried out. Compared to the standard LMQ, our new approach achieves a signal-to-noise ratio (SNR) gain of about 1 dB both in error-free and error-prone transmission conditions.
Han, S.; Fingscheidt, T
in Proc. of IEEE Int. Conf. on Acoustics, Speech, and Signal Processing (ICASSP), Florence, Italy, May. 2014
Variable-length codes (VLCs) are widely used in media transmission. Compared to fixed-length codes (FLCs), VLCs can represent the same message with a lower bit rate, thus having a better compression performance. But inevitably, VLCs are very sensitive to transmission errors. In this work, based on the trellis representation for VLCs and the BCJR algorithm, we present a variable-length softdecision decoder utilizing bit-wise channel reliability information and achieving a better error robustness in contrast to hard-decision decoding. Given the application of VLCs in audio coding showing both source correlation and variable block lengths, a strong dependency of performance is observed for both. Therefore, we point out tradeoffs of (soft-decision) decoded FLCs and VLCs depending on quantization bit rate, source correlation, and block length. We find that VLCs over AWGN channels are only recommended for very low source correlation in combination with very short block lengths and soft-decision decoding.
Jahrgang 1966 | |
1987-1993 | Studium der Elektrotechnik an der RWTH Aachen (mit Auszeichnung) |
1994 | Auszeichnung mit dem Springorum-Preis der RWTH Aachen |
1993-1998 | Wissenschaftlicher Mitarbeiter am Institut für Nachrichtengeräte und Datenverarbeitung. Arbeitsgebiet: Quellen- und Kanalcodierung, Sprachverarbeitung, Fehlerverdeckung nach mobiler Übertragung |
1998 | Promotion (mit Auszeichnung) |
1999 | Verleihung der Borchersplakette der RWTH Aachen |
1998-1999 | AT&T Labs (Shannon Labs), Florham Park, N.J., USA. Scientific Consultant. Arbeitsgebiet: Gemeinsame Sprach- und Kanalcodierung |
1999 | Auszeichnung mit dem Förderpreis der Mannesmann (jetzt Vodafone) Mobilfunkstiftung |
1999-2005 | Siemens Communications, Mobile Phones, München, zuletzt Leiter der Gruppe "Digitale (Sprach-)Signalverarbeitung" (F&E) |
2002 | Verleihung des Preises der Informationstechnischen Gesellschaft (ITG) im VDE |
2004 | Ernennung zum Senior Member des IEEE |
2005-2006 | Siemens Corporate Technology, München, Kompetenzfeldleiter "Sprachtechnologien" |
2006-heute | Universitätsprofessor an der Technischen Universität Braunschweig und Leiter der Abteilung "Signalverarbeitung für die Verkehrsinformationstechnik" am Institut für Nachrichtentechnik. |
2008-2010 | Associate Editor der Fachzeitschrift IEEE Transactions on Audio, Speech, and Language Processing |
2011 | Ernennung zum Mitglied des IEEE Speech and Language Processing Technical Committee |
2014 | Technologietransferpreis 2014 der IHK Braunschweig gemeinsam mit zwei wiss. Mitarbeitern |
Aktivitäten in Standardisierungsgremien und Verbänden:
- Seit 2007: ITU-T Fokusgruppen FITCAR und CARCOM; Mitarbeit an den ITU-T-Empfehlungen P.1100 und P.1110 für Qualitätstests von automotiven Freisprechsystemen.
- Seit 2008: Local Liaison Officer der EURASIP (European Association for Signal Processing)
- Seit 2010: Vertrauensdozent der Friedrich Naumann-Stiftung für die Freiheit, Potsdam
Gesamtliste der Veröffentlichungen
