Fractional Fourier Transformation-Based Blind Chromatic Dispersion Estimation for Coherent Optical Communications

Abstract

In this paper, we propose and demonstrate a blind chromatic dispersion (CD) estimation method based on fractional Fourier transformation (FrFT). Through numerical simulations, the proposed CD estimation method is shown to be robust against amplified spontaneous emission (ASE) noise and nonlinear interference. Only 2048 samples are required for reliable CD estimation for single carrier 28 GBaud DP-QPSK or 32 GBaud DP-16QAM signals and the standard deviation can be as low as 98.9 ps/nm and 103.6 ps/nm respectively. The feasibility of the proposed CD estimation method has been experimentally verified using 28 GBaud DP-QPSK and 14 GBaud DP-16QAM signals over various transmission distances with CD ranges of 46332 ps/nm ~ 77220 ps/nm and 5148 ps/nm ~ 51480 ps/nm respectively. Compared with some other CD estimation methods, the method based on FrFT has advantages in the aspects of less computation complexity and robustness to transmission impairments.

Joint Modulation Format Identification and Frequency Offset Estimation Based on Superimposed LFM Signal and FrFT

“…In our simulations, p TO is set to be 0.05 and maximum TO is ±4096, so the TO induced maximum peak deviation is ±322 according to Eq. (17). That means the FO estimation range is [−0.4818R s , +0.4818R s ] when a TO of 4096 symbols is in existence, which is also close to the theoretical interval.…”

Estimation of Chromatic Dispertion with FR. FT in Optical Cable

“…Since order is (2* α) / pi =0 implies Here C is the LFM parameter and signal can be visualized with an AF [1] in the guided media. So the guided wave represented as following equation [2] = M. e t2 7So f(u)= dt 8If So order p can be (9) Then the order can be obtained as in (10) with subsisting in [9]. (10)…”

Joint Time/Frequency Synchronization and Chromatic Dispersion Estimation With Low Complexity Based on a Superimposed FrFT Training Sequence

“…Two groups of TS1 & TS2 and TS3 & TS2 have opposite orders, by which two groups of TOEs and FOEs can be obtained similar to [5]. In [5] the preset frequency of two opposite orders TS are both 0, so TOE and FOE should be revised due to preset FO compared with (5) and (6): Where ± represents − when i = 1; ± represents + when i = 3. Due to preset offset, CD will influence TO between TS1 & TS2 and TS3 &TS2.…”

“…We average 2 FOEs from (13) as the final FOE value. Then the rough CDE from (9) and precise CDE in [6] are followed. Our TOE method might have several samples error, so a known synchronization TS is inserted to locate the frame head by correlation point by point.…”

“…However, the FOE resolution in [5] is not precise enough and requires fractional FOE further. In addition, an FrFT based CDE method [6] has a stronger robustness to poor channel conditions and relatively lower computation complexity compared with [3] and [4]. However, it is still a CDE method by cost function scanning, in which every scan requires one time FrFT and auto-correlation.…”