透過您的圖書館登入
IP:3.17.183.24
  • 學位論文

基於非二位元的時變累加碼之渦輪碼設計與分析

The Design and Analysis of Turbo Codes Based on Nonbinary Time-Varying Accumulate Codes

指導教授 : 邱茂清
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


文獻上已經證明基於belief-propagation(BP)解碼演算法下的隨機旁集(random-coset)非二位元GF(q)的低密度檢查碼(low-density -parity-check code, LDPC codes)及非規則重複累加碼(irregular repeat-accumulate code, IRAcode)搭配q-元非均勻訊號星雲可以達到無限制的通道極限。隨機旁集GF(q)LDPC碼的編碼複雜度比IRA碼還高,原因是IRA碼可以使用簡單的時變累加碼(time-varying accumulate code)進行編碼。初步模擬顯示隨機旁集GF(q)的LDPC及IRA碼,當變數節點階數固定為2時就可達到相當好的性能。這樣的結果意味著對於非二位元GF(q)重複壘加碼(repeat-accumulate code, RA code),只要將訊息重複兩次就可得到相當好的性能。因為訊息重複兩次,該編碼器就會類似於渦輪碼(turbo code)。因此我們預測如果將時變累加碼取代原有渦輪碼中的迴旋碼(convolutional code)應該可以得到相當不錯的性能。這就是這篇論文的研究動機。隨後我們會專注於基於非二位元的時變累加碼之渦輪碼,並利用外在訊息轉換圖(extrinsic information transfer(EXIT) charts)的技術來分析及設計好的碼,並與GF(q)下的不規則重複累加碼和規則重複累加碼做比較。

並列摘要


It has been shown that, under belief-propagation (BP) decoding, random-coset GF(q)low-density parity-check (LDPC) codes and irregular repeat-accumulate (IRA) codes with q-ary nonuniform signal constellations approach the unrestricted Shannon limit. Random-coset GF(q)LDPC code has much higher encoding complexity than that of the IRA code, since the IRA code can be encoded using the structure of time-varying accumulate code [5]. However, we observed that best SNR thresholds of random-coset GF(q)LDPC or IRA codes are obtained for average variable node degree as 2. This means that for repeat-accumulate (RA) codes twice repetitions may be sufficient to provide good performance which implies that turbo codes with two branches encoded by two independent time-varying accumulate codes may be a good alternative to construct good codes. This is the research motivation of this paper. Then we will concentrate our attention on the Turbo code based on GF(q) time-accumulated code, and use the extrinsic information transfer(EXIT) charts to design and make good codes. Then we use our simulation result to compare with the GF(q) irregular repeat-accumulate (IRA) codes.

參考文獻


[2] A. Bennatan and D. Burshtein, “Design and analysis of nonbinaryLDPC codes for arbitrary discrete-memoryless channels,” IEEE Trans.Inf. Theory, vol. 52, no. 2, pp. 549–583, Feb. 2006.
[4] K. Yang, “A nonbinary extension of RA codes: weighted nonbinary repeataccumulate codes,” in Proc. 14th IEEE Int. Symp. Personal IndoorMobile Radio Commun., Beijing, China, Sep. 2003, pp. 2882–2885
[5] Mao-Ching Chiu,“Bandwidth-Efficient Modulation Codes Based onNonbinary Irregular Repeat Accumulate Codes,” IEEE Digital Object Identifier:10.1109/GLOCOM.2008.
[6] M.-C. Chiu, ”Bandwidth-efficient modulation codes based on nonbinary irregular repeat-accumulate codes,” IEEE Trans. Inf. Theory, vol. 56, pp. 152–167, Jan. 2010.
[7] C. Berrou, A. Glavieux, P. Thitimajshima, "Near Shannon limit error-correcting coding and decoding: Turbo Codes," Proc. IEEE Intl. Conf. Commun., pp. 1064-70, Geneva, Switzerland, May 1993.

延伸閱讀