BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A Cross-Entropy Based Method to Analyse Iterative Decoding - Qingl
 in Luo
DTSTART:20051012T140000Z
DTEND:20051012T150000Z
UID:TALK4430@talks.cam.ac.uk
CONTACT:Phil Cowans
DESCRIPTION:Iterative decoding provides a practical solution for the appro
 aching of Shannon limit with acceptable complexity. By decoding in an iter
 ative fashion\, the decoding complexity is spread over time domain while t
 he overall optimality is still approachable.\n \nEver since its successful
  application in turbo codes in 1993\, people keep trying to discover the s
 ecrets behind iterative decoding. Till now\, BER bounds\, density evolutio
 n\, EXIT chart\, Gaussian approximation are several most famous methods th
 at proves to be helpful for the analysis of the behavior of iterative deco
 ders. However\, restrictions like subject sequence must be Gaussian distri
 buted\, transmitted sequence must be known\, applicable region is either B
 ER floor or BER clip only\, etc.\, greatly limit the applications of these
  methods.\n\nIn this talk\, a new\, universal method for the analysis of i
 terative decoding based on cross-entropy will be discussed. We prove that 
 the maximum a posteriori probability (MAP) decoding algorithm minimizes th
 e cross-entropy between the a priori and the extrinsic information subject
  to given coding constraints\, and the error correcting ability of each st
 ep of decoding can be evaluated with this cross-entropy for a converging t
 urbo decoder. Based on this proof\, the analysis of turbo decoding on conv
 ergence rate\, derivation of Eb/N0 convergence threshold\, evaluation of e
 rror performance in "error floor" region\, and design of asymmetric turbo 
 codes are carried out. Unlike most conventional analysis methods which rel
 y heavily on either Gaussian approximation of distribution of the a priori
 /extrinsic information or a full knowledge of source bits\, or even both\,
  the new method provides analysis in a totally blind fashion.
LOCATION:HEP Seminar Room\, Cavendish Laboratory
END:VEVENT
END:VCALENDAR
