Performance analysis of short analog fountain codes

Abbas, Rana, Shirvanimoghaddam, Mahyar, Huang, Tao, Li, Yonghui, and Vucetic, Branka (2019) Performance analysis of short analog fountain codes. In: Proceedings of the 2019 IEEE Globecom Workshops. From: 2019 IEEE Globecom Workshops, 9-13 December 2019, Waikoloa, HI, USA.

[img] PDF (Published Version) - Published Version
Restricted to Repository staff only

View at Publisher Website: https://doi.org/10.1109/GCWkshps45667.20...
 
4


Abstract

Analog fountain codes are a class of rateless codes that have been demonstrated to achieve near-capacity performance for asymptotically long blocks, without any channel state information at the transmitter side. Recently, a new design for these codes has been proposed aimed at improving its performance in the finite block length regime, dubbed short analog fountain codes (S-AFC). S-AFC was shown to score error rates orders of magnitude smaller than AFC for blocks of a few hundred bits long. S-AFC was also shown to achieve average block lengths close to the Polyanskiy-Poor and Verdu bound for high SNR and exhibits no error floors down to 10^-7. In this paper, we derive lower and upper bounds on the block error rate (BLER) of S-AFC. We verify these bounds through Monte Carlo simulations and provide all the proofs in the appendix.

Item ID: 62715
Item Type: Conference Item (Research - E1)
ISBN: 978-1-7281-0960-2
Related URLs:
Copyright Information: Copyright © 2019 by the Institute of Electrical and Electronics Engineers,Inc.
Date Deposited: 09 Jul 2020 02:10
FoR Codes: 10 TECHNOLOGY > 1005 Communications Technologies > 100504 Data Communications @ 50%
10 TECHNOLOGY > 1005 Communications Technologies > 100510 Wireless Communications @ 50%
SEO Codes: 89 INFORMATION AND COMMUNICATION SERVICES > 8901 Communication Networks and Services > 890103 Mobile Data Networks and Services @ 50%
89 INFORMATION AND COMMUNICATION SERVICES > 8901 Communication Networks and Services > 890104 Mobile Telephone Networks and Services @ 50%
Downloads: Total: 4
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page