Recent results on NB-LDPC decoder and Noise against noise decoder
Event details
| Date | 25.01.2017 |
| Hour | 11:15 |
| Speaker | Prof. Emmanuel Boutillon, Lab-STICC, Université de Bretagne Sud |
| Location | |
| Category | Conferences - Seminars |
In this two fold presentation, we will first present new results on Non Binary Low Density Parity Check (NB-LDPC) code. NB-LDPC codes are error correcting codes that outperform the binary error control codes proposed for the next 5G communication standard. The drawback of NB-LDPC codes is their higher hardware decoding complexity. Some recent results on simplified hardware architecture implementation will be presented.
The second part of the talk will be dedicated to the benefits of noise in error control decoders. In natural science, noise is sometimes recognized as a source of innovation and improvement (erroneous copy of DNA in evolution, micro-saccade of eyes for high resolution vision). In the signal processing community, the idea of using noise to resolve some particular detection and estimation problems is rather old (stochastic resonance). In spite of its potential, “noised aided signal processing” stays marginal. In this presentation we will show how noise can help a simple decoding algorithm to decode efficiently Low Density Parity Check (LDPC) Code. Hardware implementation of a "noise assisted" decoder can compete in terms of area, power dissipation and performance with the best known classical hardware implementation.
Bio: Emmanuel Boutillon received the Engineering Diploma from the Telecom Paris Tech, Paris, France in 1990. In 1991, he worked as an assistant professor in the Ecole Multinationale Supérieure des Télécommunications in Africa (Dakar). In 1992, he joined Telecom Paris Tech as a research engineer where he conducted research in the field of VLSI for digital communications. He obtained his Ph.D in 1995. In 1998, he spent a sabbatical year at the University of Toronto, Ontario, Canada. In 2000, he moved to the University of Bretagne Sud as a professor. He headed the LESTER lab from 2005 up to end of 2007 and the CACS department between 2008 up to 2015. He is currently the scientific adviser of the lab-STICC. In 2011, he had a sabbatical year at INICTEL-UNI, Lima (Peru).
His research interests are on the interactions between algorithm and architecture in the field of wireless communications and high speed signal processing. In particular, he works on Turbo Codes and LDPC decoders.
The second part of the talk will be dedicated to the benefits of noise in error control decoders. In natural science, noise is sometimes recognized as a source of innovation and improvement (erroneous copy of DNA in evolution, micro-saccade of eyes for high resolution vision). In the signal processing community, the idea of using noise to resolve some particular detection and estimation problems is rather old (stochastic resonance). In spite of its potential, “noised aided signal processing” stays marginal. In this presentation we will show how noise can help a simple decoding algorithm to decode efficiently Low Density Parity Check (LDPC) Code. Hardware implementation of a "noise assisted" decoder can compete in terms of area, power dissipation and performance with the best known classical hardware implementation.
Bio: Emmanuel Boutillon received the Engineering Diploma from the Telecom Paris Tech, Paris, France in 1990. In 1991, he worked as an assistant professor in the Ecole Multinationale Supérieure des Télécommunications in Africa (Dakar). In 1992, he joined Telecom Paris Tech as a research engineer where he conducted research in the field of VLSI for digital communications. He obtained his Ph.D in 1995. In 1998, he spent a sabbatical year at the University of Toronto, Ontario, Canada. In 2000, he moved to the University of Bretagne Sud as a professor. He headed the LESTER lab from 2005 up to end of 2007 and the CACS department between 2008 up to 2015. He is currently the scientific adviser of the lab-STICC. In 2011, he had a sabbatical year at INICTEL-UNI, Lima (Peru).
His research interests are on the interactions between algorithm and architecture in the field of wireless communications and high speed signal processing. In particular, he works on Turbo Codes and LDPC decoders.
Practical information
- General public
- Free
Organizer
- Prof. ANDREAS P. BURG Tel: +41-21-693 69 24 Director of the Telecommunications Circuits Laboratory
Contact
- Prof. ANDREAS P. BURG Tel: +41-21-693 69 24 Director of the Telecommunications Circuits Laboratory