應用於光通訊的受限選擇軟性RS 解碼器

Abstract

在光通訊系統中，因為更加快速的傳輸速度而增加的訊號不穩定性，錯誤更正裝 置需要提供更加有力的錯誤更正能力。相較於傳統硬性RS解碼器，軟性解碼器 可以提供更好的錯誤更正能力，但也需要高出許多的硬體複雜度。本論文提出了 一種受限選擇軟性演算法不但可以增加錯誤更正能力並且保有面積效益。創新的 重點在於，不是將許多可能的傳輸訊息解碼再從中挑選出最有可能的一組，而是 將錯誤位置數學式Λ(x)設下限制導致只有解碼其中一組。根據RS(255,239)的模擬 結果，在10−4 CER時可比硬性RS解碼器多達0.4 dB傳輸效益。實驗結果顯示我們 提出的軟性解碼器在CMOS 90奈米製程下可達到2.56 Gb/s吞吐量並且只需花費 與硬性解碼器相似的硬體複雜度。此軟性解碼器可以適用於利用16套解碼器的 10-40 Gb/s光纖系統以及2.5 Gb/s GPON等應用。

Due to the increasing uncertainty of data for higher transmission rate, the Forward Error Correction (FEC) devices need to provide more powerful error correcting capability for optical communication systems. As compared with traditional hard RS decoders, the soft RS decoders can perform substantial coding gain but require much higher hardware complexity. In this thesis, a decision-confined algorithm is proposed to enhance the error correcting performance with an area-efficient architecture. The novelty is that, instead of decoding numerous possible transmitted codewords and choosing the most likely one, only one candidate sequence will be decoded after confining the degree of error-locator polynomial Λ(x). For RS (255,239) codes, simulation results confirm that our approach provides 0.4 dB performance gain at 10−4 CER over the hard RS decoders. The experimental result reveals that our soft decoder can achieve 2.56 Gb/s throughput in standard CMOS 90 nm technology while having similar complexity as a hard decoder. It can fit well for 10-40 Gb/s with 16 RS decoders in optical fiber systems and 2.5 Gb/s GPON applications.