Summary This study investigated ten polyphenolic compounds in the peel of 145 apple cultivars grown in the same location while originated from different countries over two successive years. Despite significant year effect of individual and total polyphenol concentrations on each cultivar, both concentrations showed similar distributions in the 2 years, and genotypic correlations among them were relatively stable. Genotypic variations were considerable, with total polyphenol concentration ranging from 363.9 to 2516.9 μg g−1 FW. Cyanidin‐3‐galactoside was found only in red apple peels, at levels of 67.8–371.6 μg g−1 FW. Flavanols (epicatechin and catechin) and dihydrochalcones (mainly phloridzin) accounted for 8.8–66.1% and 8.58–61.2% of total polyphenols, respectively. Six flavonols (quercetin‐3‐galactoside, quercetin‐3‐glucoside, quercetin‐3‐rhamnoside, quercetin‐3‐xyloside, quercetin‐3‐arabinoside, and quercetin‐3‐rhamnoglucoside) were identified. The proportion of flavonols was 9.3–70.6% of total polyphenols. This wide variation could help in selecting cultivars with different polyphenol patterns in the peel. Principal component analysis showed that cultivars originated from different countries could not be distinguished in terms of polyphenol profiles.

We present a novel method to estimate chromatic dispersion with ultra-low sampling rate based on training sequence. The under-sampling can be equivalent to oversampling by splicing points with periodic sequence. The functional relationship between equivalent over-sampling rate and symbol rate, actual sampling rate and sequence period is deduced. To demonstrate the feasibility of the method, the simulation of a 28 GBaud QPSK optical fiber communication system is carried, in which 76 MSa/s sampling rate is equivalent to 66.5 GSa/s sampling rate. The results show that maximum estimation error is less than 160 ps/nm after 100 km to 2300 km SSMF transmission. We also demonstrate the robustness of the proposed method to amplified spontaneous emission and nonlinear noise in a three-channel optical fiber communication system. Moreover, the proposed method has been experimentally verified with a 20.5 GBaud QPSK system, 1.25 GSa/s sampling rate is equivalent to 51.25 GSa/s sampling rate, and estimation error is less than 100 ps/nm and 150 ps/nm after 100 km and 200 km SSMF transmission, respectively. The proposed method eliminates the requirement of high speed ADC and is cost effective, which can be used for optical performance monitoring.

We design a quadrature phase-shift keying training sequence (QPSK-TS) for monitoring both optical signal-to-noise ratio (OSNR) and chromatic dispersion (CD) in a dense wavelength division multiplexing (DWDM) system. By applying fast Fourier transform on the QPSK-TS, a comb-like spectrum with discrete peaks can be obtained. OSNR can be calculated by measuring the power of the amplified spontaneous emission (ASE) noise in the gaps between these peaks. Moreover, we search the optimal order in the fractional domain to calculate the CD after applying fractional Fourier transform on the QPSK-TS signal. The proposed method shows large ASE noise and optical nonlinearity tolerance in simulation results (<1.0 dB error within 24 dB after transmission over 1000 km with 3 dBm per channel launch power, and <90 ps/nm error within 33400 ps/nm). Furthermore, the experimental results demonstrate high accuracy of the proposed method (≤1.0 dB error within 28 dB and <40 ps/nm error within 1670 ps/nm after 100 km single channel transmission).

Fast and low-cost chromatic dispersion (CD) estimation is very important for fiber link that is dynamically reconfigurable in the next-generation optical networks. In this paper, a novel CD estimation method based on the deep neural network (DNN) with ultra-low sampling rate is proposed for the optical fiber transmission systems. The method can estimate CD for under-sampled signals by the trained DNN with a weighted average. To demonstrate the feasibility of the method, the simulation of a 28-GBaud fivechannel optical fiber transmission system with 100-2000-km SSMF transmission is carried, where the sampling rate of the analog-to-digital converter (ADC) at the receiver is only 500 MHz. The results show that the maximum mean absolute error (MAE) of the estimated CD is less than 75 ps/nm with the reference cumulative CD from 1600 to 32 000 ps/nm for QPSK and 16QAM signals. Meanwhile, the robustness of the proposed method is verified for amplifier spontaneous emission (ASE) and nonlinear (NL) noise. Furthermore, we experimentally demonstrated that the maximum MAE is less than 75 ps/nm when the transmission distance varies from 100 to 800 km in 20-GBaud QPSK optical fiber transmission system. In conclusion, the proposed DNN-based CD estimation method shows great potential for the cost-effective under-sampled systems. INDEX TERMS Metrology, fiber optics, fiber nonlinear optics, chromatic dispersion.

No abstract