LDPC碼的位元翻轉解碼的改進; Improved Bit-Flipping Decoding of Low-Density Parity-Check Codes

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Communication Engineering > Electronic Thesis & Dissertation > Item 987654321/8183

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/8183

Title:	LDPC碼的位元翻轉解碼的改進;Improved Bit-Flipping Decoding of Low-Density Parity-Check Codes
Authors:	張書瑋;Shu-wei Zhang
Contributors:	通訊工程研究所
Keywords:	錯誤更正碼;LDPC;bit-flipping
Date:	2007-06-25
Issue Date:	2009-09-22 11:20:22 (UTC+8)
Publisher:	國立中央大學圖書館
Abstract:	利用訊息傳遞解碼演算法，我們發現在解碼方面有很好的效能，但因為其複雜度偏高，所以，很多研究都在討論如何改善複雜度，又不會使效能嚴重的衰減。相對於硬式解碼演算法在效能方面不像軟式解碼演算法這麼出色，因此，較少的研究提出來改善硬式解碼演算法。不過，硬式解碼演算法也有優點，像是複雜度低、易於硬體實現。在這篇論文，主要是要討論低複雜度的解碼演算法，因為每次疊代只更正ㄧ個位元節點，所以，需要多次疊代才能完成解碼動作。因此，在這裡提出ㄧ個方法，就是利用位元節點間的不相關性，判斷哪些位元節點是要翻轉，並且隨著疊代越多次，表示位元節點越來越可靠，不可任意翻轉，有一機制來判定是否要翻轉多位元節點。在每次疊代翻轉多個位元節點，可以發現疊代次數在較少的情況下，每次疊代翻轉多個位元節點的效能比每次疊代翻轉一個位元節點好很多。當要達到相同的效能時，每次疊代翻轉多個位元節點，可使用較少的疊代數目即可達到，因此，可以有效的節省解碼時間，提高解碼速度。 Utilize the message-passing decoding algorithm , we find that there is very good performance in decoding. But, because its complexity is on the high side, a lot are studying and discussing how to improve complexity , and it will not make the performance decay seriously. Decode algorithm of performing in efficiency as to hard type as soft to is it perform algorithm so outstanding to decode. So, less research puts forward and improves the performance of hard-decision decoding algorithms. However, the hard-decision decoding algorithms have some advantages. as if low complexity, and realizing of hardware easily. In this thesis , I will mainly discuss decoding algorithms of low- complexity, because it corrects one bit-node per iteration. so, it needs many iterations to finish decoding movements . Hence, I propose a method here. Utilizing the non-correlation of bits nodes and than judging what bit nodes to be flipped. As increasing the frequency of iterations, it represents the bit nodes are more reliable and can not be flipped arbitrarily. It should have a mechanism to judge if flipping multi-bits. When flipping multi-bits per iteration, we can find the performance of flipping multi-bits per iteration better than flipping one bit per iteration in the condition of less iterations. Flipping multi-bits per iteration can use less iterations to achieve the same performance. Therefore , it can save decoding time effectively and improve the speed of decoding.
Appears in Collections:	[Graduate Institute of Communication Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Size	Format

社群 sharing

Loading...