Scheduled System Maintenance:
On Monday, April 27th, IEEE Xplore will undergo scheduled maintenance from 1:00 PM - 3:00 PM ET (17:00 - 19:00 UTC). No interruption in service is anticipated.
By Topic

A flexible NoC-based LDPC code decoder implementation and bandwidth reduction methods

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)

The need for efficient and flexible LDPC (Low Density parity Check) code decoders is rising due to the growing number and variety of standards that adopt this kind of error correcting codes in wireless applications. From the implementation point of view, the decoding of LDPC codes implies intensive computation and communication among hardware components. These processing capabilities are usually obtained by allocating a sufficient number of processing elements (PEs) and proper interconnect structures. In this paper, Network on Chip (NoC) concepts are applied to the design of a fully flexible decoder, capable to support any LDPC code with no constraints on code structure. It is shown that NoC based decoders also achieve relevant throughput values, comparable to those obtained by several specialized decoders. Moreover, the paper explores the area and power overhead introduced by the NoC approach. In particular, two methods are proposed to reduce the traffic injected in the network during the decoding process, namely early stopping of iterations and message stopping. These methods are usually adopted to increase throughput. On the contrary, in this paper, we leverage iteration and message stopping to cut the area and power overhead of NoC based decoders. It is shown that, by reducing the traffic injected in the NoC and the number of iterations performed by the decoding algorithm, the decoder can be scaled to lower degrees of parallelism with small losses in terms of BER (Bit Error Rate) performance. VLSI synthesis results on a 130 nm technology show up to 50% area and energy reduction while maintaining an almost constant throughput.

Published in:

Design and Architectures for Signal and Image Processing (DASIP), 2011 Conference on

Date of Conference:

2-4 Nov. 2011