基於渦輪碼之分交錯器多重存取

Abstract

分交錯器多重接取 (IDMA) 乃利用不同的交錯器 (interleaver) 來辨別用戶的一種多重接取技術。此技術可利用簡單的碼片 (chip by chip, CBC) 多用戶偵測方法，讓偵測用戶的計算複雜度不會因用戶數的增加而遽增。相較於CDMA需事先設計具低相關性展頻碼群才能有較好的效能，IDMA使用具一定亂度的隨機產生的交錯器即可達到不錯的效能，因此用戶容量相對的高。在本論文中，我們將渦輪碼與IDMA系統結合，比較了利用渦輪碼編碼與傳統的捲積碼的IDMA系統在不同參數條件下效能。此外，我們也將原本使用BPSK之調變法改成DPSK調變，讓重複器(repetition unit)與交錯器加DPSK調變組成一等效的重複累積碼(repetition-accumulate code, RA code)並加上額外的交錯器。從電腦模擬結果可發現相較於使用捲積碼之傳統IDMA系統，使用渦輪碼編碼之IDMA系統能有較佳的效能；另外使用DPSK調變之IDMA系統，其效能亦比其他兩種IDMA系統有較低的錯誤層(error floor)。

Interleave division multiple access (IDMA) is a multiple access scheme that employs user-specific interleavers to differentiate users. It uses a simple chip by chip multiuser detection method which requires a constant per-user computational complexity, independent of number of system users. Compared with the coded division multiple access (CDMA) method, IDMA scheme can provide single-user performance by using simple repetition coding and random interleaving in addition to a common forward error correcting (FEC) code while CDMA requires a family of low cross-correlation spreading codes. In this thesis, we first consider IDMA systems that use either a turbo or convolutional code as the common FEC code. An alternative system structure replaces BPSK modulator by a DPSK modulator and insert an interleaver after the DPSK modulator. The combined repletion (unit) interleaver and DPSK modulator is equivalent to a repetition-accumulation (RA) encoder. Numerical results show that our turbo code based IDMA (TC-IDMA) scheme is able to outperform the conventional (convolutional code based) IDMA system and the DPSK-based system yields much lower error floor.