Talk:Luby transform code

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Forward Error Correction code[edit]

LT code are Forward error correction codes, there is no mention of this in the article! In "Why use an LT code?" we can read that rateless code are the solution for the transmission over an erasure channel with no feedback channel. Rateless codes are not the only solution to this problem. Any Forward Error Correction code can do the job (with an appropriate code rate with regard to the loss probability of the channel). --Cunchem 13:09, 15 April 2009 (UTC) —Preceding unsigned comment added by Cunchem (talkcontribs)

This article doesn't make sense[edit]

I read this article three times, and I still can't really get the flavor of LT codes out of it. I think it needs a major rewrite.

From reading other literature, it appears that LT codes depend entirely on the Exclusive OR operation, or bitwise addition with no carry out of any bit position. I intend to rewrite the description of LT codes to make this clear to a typical Wiki reader. Probably tomorrow. DavidCBryant 00:44, 26 February 2007 (UTC)

Comment:[KiMoon Lee] I think, the excercises in D. Mackay's book explains the Robust Soliton Distribution within one page (see chapter 50 in Mackay's Inference book). I hope this could help you.
Answer: [Ray GAO] [1] this is david mackay's explaination to LT code. It basically starts from Linear Fountain Codes. It should solve your problem

Patent issues[edit]

M. Luby. Information additive code generator and decoder for communication system. U.S. Patent #6,373,406. Issued April 16, 2002. (continuntion to #6,307,487).

This patent seems to describe LT codes. Digital Fountain now have the patent rights for the aforementioned patent. Digital Fountain also have a patent on Raptor codes. Raptor codes use LT codes after encoding with a systematic code, such as Hamming or LDPC codes. For patent infringement to occur, all claims listed in the patent must be replicated in a commercial service or product. Does anyone want to correct these statements? —The preceding unsigned comment was added by 137.215.9.20 (talk) 08:26, 8 May 2007 (UTC).