Title :
Polar codes are optimal for lossy source coding
Author :
Korada, Satish Babu ; Urbanke, Rüdiger
Author_Institution :
EPFL, Lausanne, Switzerland
Abstract :
We consider lossy source compression of a binary symmetric source with Hamming distortion function. We show that polar codes combined with a low-complexity successive cancellation encoding algorithm achieve the rate-distortion bound. The complexity of both the encoding and the decoding algorithm is O(N log(N)), where N is the blocklength of the code. Our result mirrors Ar¿kan´s capacity achieving polar code construction for channel coding.
Keywords :
channel coding; computational complexity; rate distortion theory; Hamming distortion function; binary symmetric source; channel coding; decoding algorithm; lossy source coding; polar codes; Belief propagation; Channel coding; Decoding; Hydrogen; Information theory; Iterative algorithms; Nonlinear distortion; Parity check codes; Rate-distortion; Source coding;
Conference_Titel :
Information Theory Workshop, 2009. ITW 2009. IEEE
Conference_Location :
Taormina
Print_ISBN :
978-1-4244-4982-8
Electronic_ISBN :
978-1-4244-4983-5
DOI :
10.1109/ITW.2009.5351488