FXFei Xie

Fei Xie

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Fei Xie has a total of 28 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Training Sequence and Amplitude Modulation Pilot Tone Aided Nonlinear Interference Monitoring for Coherent Optical Communication Systems

    by Hao Zhou, Wen Zuo, Fan Yang, Du Tang, Yingjie Jiang, Fei Xie, Yan Zhao, Bing Ye, Hengying Xu, Chenglin Bai, et al.

    A nonlinear interference (NLI) monitoring method using a training sequence (TS) and amplitude modulation pilot tone (AMPT) is proposed for coherent optical transmission systems. To estimate the total noise, including NLI noise and amplifier spontaneous emission (ASE) noise, the TS is used to obtain prior information and subtract it from the received signal. Then, to separate the NLI noise, including self-channel interference (SCI) and cross-channel interference (XCI) noise, from the total noise, the AMPT induces power oscillations in the NLI noise so that the NLI and ASE powers can be extracted using different frequency characteristics. Finally, the NLI-induced optical SNR (OSNR) penalty can be estimated using these powers. Verified by a 5-channel 64-GBaud 1500-km simulation system, the proposed method can accurately estimate the SNR of total, SCI, XCI, and ASE noise, achieving NLI-induced OSNR penalty estimation errors of less than 0.39 dB. We also conducted an experiment in a 64-GBaud 500-km coherent transmission system with dual-polarization 16 quadrature amplitude modulation (DP-16QAM). The OSNR penalty estimated by the proposed method achieves a 0.41-dB monitoring error with 50-$\mu$s time-window data. The proposed method offers a better solution for achieving low-cost, real-time, and high-accuracy fiber-optical performance monitoring in coherent transmission systems.

  • Pilot symbols and amplitude modulation pilot tone aided nonlinear interference monitoring for coherent systems

    by Hao Zhou, Wen Zuo, Du Tang, Fan Yang, Yingjie Jiang, Yutao Zhou, Fei Xie, Baoxuan Quan, Huan Chen, Yan Zhao, et al.

    A nonlinear interference (NLI) monitoring method is proposed for coherent transmission systems by jointly using implementation agreement-defined pilot symbols (PS) and amplitude modulation pilot tone (AMPT). The PS is employed to extract the total noise from the received signal, while the AMPT enables separation of self-channel interference (SCI) and cross-channel interference (XCI) power from amplifier spontaneous emission (ASE) power. In a 60-GBaud dual-polarization quadrature phase-shift keying (DP-QPSK) 1500-km wavelength-division multiplexing (WDM) simulation system, the NLI signal-to-noise ratio (SNR) and NLI-induced optical SNR (OSNR) penalty were estimated with errors less than 0.26/0.07 dB for a single-channel (SCI-only) system and 0.52/0.30 dB for a 5-channel (SCI+XCI) system within a ∼160-µs time window. In a single-channel 60-GBaud DP-QPSK 500-km experimental system, the estimation errors of NLI SNR and OSNR penalty were below 0.45 dB and 0.35 dB, respectively, using ∼ 300-µs data. The proposed PS-AMPT NLI monitoring method exhibits high accuracy and strong compatibility with deployed optical networks.

  • Low-complexity symbol-rate equalization schemes for short-reach coherent-lite optical interconnects

    by Baoxuan Quan, Fei Xie, Hao Zhou, Fan Yang, Yutao Zhou, Yingjie Jiang, Chenglin Bai, Hengying Xu, Du Tang, Yaojun Qiao

    In this paper, we propose two low-complexity multiple-input multiple-output (MIMO) equalization schemes for short-reach coherent-lite optical interconnects: dynamic pruning MIMO (DP-MIMO) and dynamic clustering MIMO (DC-MIMO). Based on the conventional 4 × 4 real-valued (RV) MIMO architecture, both schemes aim to reduce the computational complexity of each channel filter according to real-time channel characteristics. DP-MIMO sparsifies filter coefficient matrices and dynamically adjusts the sparsity level of each channel filter based on the energy distribution across all channels, achieving adaptive tap optimization. DC-MIMO reduces redundant computations by clustering filter coefficients and further introduces an energy-adaptive mechanism that dynamically adjusts the number of clusters, balancing complexity and system performance. To further evaluate the benefits of the energy-adaptive mechanism, the static pruning MIMO (SP-MIMO) and static clustering MIMO (SC-MIMO) are also introduced as their static counterparts. The proposed schemes are experimentally validated in a C-band 80-GBaud dual-polarization 16QAM (DP-16QAM) system over 1-km, 2-km, and 5-km standard single-mode fiber (SSMF) links. Experimental results show that, compared with conventional 4 × 4 RV MIMO, DP-MIMO can achieve around 40% reduction in real-valued multiplications (RMs) per transmitted symbol within a 0.20-dB receiver optical power (ROP) penalty. Compared with SP-MIMO at a similar complexity level, DP-MIMO introduces a smaller ROP penalty. DC-MIMO achieves around 60% RMs reduction within a 0.15-dB ROP penalty compared with 4 × 4 RV MIMO, and compared with SC-MIMO at similar performance, DC-MIMO achieves lower complexity. To verify the robustness of the proposed schemes against in-phase/quadrature (IQ) skew impairments, an additional 2-ps IQ skew was introduced. Both DP-MIMO and DC-MIMO maintain stable performance under this condition, with DP-MIMO reducing around 40% of complexity within a 0.2-dB ROP penalty and DC-MIMO reducing around 60% of complexity within a 0.1-dB ROP penalty.


  • A Wide-Range and Accurate Skew Estimation Method Based on Coprime Spectral-Block Training Signal in Coherent Systems

    by Langjun Liu, Hao Zhou, Baoxuan Quan, Fei Xie, Fan Yang, Du Tang, Chenglin Bai, Hengying Xu, Yaojun Qiao

  • Lightweight scale-enhanced neural network-based optimized detection for bandwidth-limited IM/DD systems

    by Fei Xie, Hao Zhou, Yingjie Jiang, Hengying Xu, Chenglin Bai, Yaojun Qiao

    In this Letter, we propose a lightweight and efficient optimized detection scheme based on a lightweight scale-enhanced neural network (LSE-NN) for receiver digital signal processing (DSP) in bandwidth-limited intensity modulation and direct detection (IM/DD) transmission systems. The proposed LSE-NN-based scheme comprises three main components: input preprocessing via 2-tap scale-enhanced low-pass filtering, an NN-based lookup table (NN-LUT), and a neural network-based log-maximum à posteriori (MAP) decoder with a fixed number of surviving states (NN-MAP). The scheme is implemented in a 122-Gbps optical IM/DD system employing 4-level pulse amplitude modulation (PAM-4) over a 2-km transmission distance, with a 10-dB bandwidth of 15 GHz. Experimental results demonstrate that the proposed LSE-NN-MAP scheme improves receiver sensitivity by 2.4 dB while reducing computational complexity by more than 98.44% compared to conventional optimized detection, achieving the 7% forward error correction (FEC) threshold. To the best of our knowledge, this is the first demonstration of a lightweight and efficient neural network architecture employed for optimized detection in IM/DD systems.


  • Low-Complexity Nonlinear Interference Estimation Using WRP1-Aided Coarse Step-Size Power Profile Estimation

    by Yingjie Jiang, Du Tang, Hao Zhou, Fan Yang, Fei Xie, Hengying Xu, Chenglin Bai, Aiying Yang, Yaojun Qiao

    As an emerging optical performance monitoring (OPM) technique, power profile estimation (PPE) has gained significant research attention for its ability to reveal the longitudinal signal power evolution along optical fibers. Among its various applications, nonlinear interference (NLI) noise estimation is particularly promising. In this paper, we propose a low-complexity NLI noise estimation method based on coarse step-size PPE combined with the weighted first-order regular perturbation model (WRP1). The WRP1 model accounts for the correlation between neighboring symbols in the nonlinear operator, thereby mitigating the inaccuracies introduced by coarse step-size power estimation and enhancing overall NLI estimation performance. The proposed method is validated through numerical simulations over a 1000 km transmission system and experimental demonstrations over 500 km and 1200 km optical links. The results demonstrate that the proposed method achieves an estimation error of less than 0.5 dB in both the nonlinear signal-to-noise ratio ($\text{SNR}_{\text{NL}}$) and OSNR penalty, which is comparable to that of conventional fine step-size approaches. Meanwhile, it reduces computational complexity by over 90%, providing a highly efficient solution for practical PPE-enabled NLI estimation.


  • High-Performance Noise Whitening Filters for Linear and Nonlinear Colored Noise Suppression in IM/DD System With Bandwidth Limitation

    by Fei Xie, Xiaoqian Huang, Chenglin Bai, Hengying Xu, Yaojun Qiao

    Although the equalizer incorporated into the digital signal processing procedure of receivers can mitigate inter-symbol interference (ISI) resulting from bandwidth limitations and chromatic dispersion, it also unavoidably enhances the noise present in the decayed frequency region. This phenomenon is known as equalization-enhanced colored noise. In this paper, we propose three high-performance noise whitening filters (NWF) for linear and nonlinear colored noise, including a scale-enhanced NWF (SE-NWF), an inverse-average NWF (IA-NWF), and an inverse-average nonlinear NWF (IA-NNWF). After obtaining a traditional linear NWF based on auto-regressive modeling, the SE-NWF incorporates an amplification factor to increase the spectral range for stronger noise-whitening ability. To further improve linear and nonlinear noise-whitening ability, the novel IA-NWF and IA-NNWF are presented, which are based on the fundamental tenet that the inversion of the noise spectrum and the extrapolation of this process through the application of linear and nonlinear finite impulse response filters, respectively. The proposed schemes are evaluated in a C-band 61-GBaud intensity modulation and direct detection (IM/DD) optical on-off keying (OOK) and 4-level pulse amplitude modulation (PAM-4) systems. In the OOK transmission with a 3.7-GHz 3-dB bandwidth over an 80-km standard single-mode fiber (SSMF), the results demonstrate that the 8-tap SE-NWF and 8-tap IA-NWF exhibit comparable performance to the 19-tap and 21-tap conventional NWFs, respectively. This indicates that the number of real-valued multiplications is reduced by 52.4% and 56.5%, respectively. With the same memory length, the SE-NWF, IA-NWF, and IA-NNWF can achieve approximate improvements of 2.8 dB, 3.0 dB, and 3.2 dB on receiver sensitivity, respectively, in comparison to the conventional NWF under a 20% forward error correction threshold. In the PAM-4 transmission over a 2-km SSMF, compared to the conventional NWF of equivalent complexity, the 2-tap proposed schemes achieve approximately 1.6 dB and 4.2 dB improvements in receiver sensitivity at the 20% and 7% FEC thresholds, respectively.


  • Advanced optimized detection scheme based on a neural network in a bandwidth-limited IM/DD system

    by Fei Xie, Xiaoqian Huang, Hengying Xu, Chenglin Bai, Yaojun Qiao

    In this Letter, we propose a high-performance optimized detection scheme based on a neural network (NN) in a receiver digital signal processing (DSP) for bandwidth-limited intensity modulation and direct detection (IM/DD) transmission systems. The NN-based optimized detection scheme consists of two components, an NN-based lookup table (NN-LUT) and an NN-based log-maximum a posteriori estimation with a fixed number of surviving state (NN-MAP) decoder. The NN-LUT provides more accurate and sufficient a priori information (PI) to the decoder than the conventional filter-form PI without increasing computational complexity. The NN-MAP decoder can be optimized for detection using the NN-form PI. The proposed scheme is implemented in a 122-Gbps IM/DD optical 4-level pulse amplitude modulation (PAM-4) system with a 3.7-GHz 3-dB bandwidth. The results demonstrate that the proposed NN-MAP scheme can enhance receiver sensitivity by 1.4 dB under the same complexity compared to conventional optimized detection, achieving a 20% forward error correction threshold. To the best of our knowledge, this represents the inaugural instance of NN being employed for optimized detection.


  • Multi-Tap Optical Delay Line-Based Optical-Electrical Dispersion Compensation Scheme in High-Speed IM/DD Systems

    by Xiaoqian Huang, Fei Xie, Wen Zuo, Yaojun Qiao

    Digital equalizer-assisted optical signal processors based on optical delay line (ODL) offers an efficient solution for chromatic dispersion (CD) compensation in intensity-modulation/direct-detection (IM/DD) transmission systems. In this paper, a multi-tap ODL-based optical-electrical dispersion compensation scheme is proposed to enhance CD compensation effectiveness. The scheme combines a low-complexity digital equalizer with multi-tap cascaded ODL (C-ODL) and feedforward ODL (F-ODL) based on distinct optical architectures. In addition, the theoretical model derivation reveals that the frequency response of the multi-tap ODL scheme achieves a flat spectrum through the superposition of diverse frequency responses. Verified in a C-band 112 Gb/s 40-km PAM4 system, C-ODL and F-ODL schemes exhibit 3-dB and 2.9-dB received optical power (ROP) gains at 7% forward error correction (FEC) limit over the single-tap ODL scheme, and achieve the KP4-FEC limit at ROPs of −5 dBm and −5.4 dBm with 30 and 53 real-valued multiplications, respectively. Furthermore, C-ODL and F-ODL schemes with optimal parameters can reach the KP4-FEC limit at the maximum CDs of 680 ps/nm and 850 ps/nm. The research results indicate that the proposed multi-tap ODL scheme exhibits significant potential for compensating CD in high-performance and complexity-effective IM/DD optical transmission systems.


  • Deployment of Ultrasmall-Size Cluster-Assisting Lookup Tables in IM/DD Systems

    by Fei Xie, Xiaoqian Huang, Shuangyue Liu, Du Tang, Chenglin Bai, Hengying Xu, Yaojun Qiao

    In the C-band high-speed long-reach optical intensity modulation direct detection (IM/DD) communication system, severe signal distortion, including chromatic dispersion and device bandwidth limitations, making the size of lookup tables (LUTs) required by equalizers too large to be operated. In this paper, we propose a cluster-assisting lookup table (CLUT) approach to reduce the table size significantly. In addition, we also propose multiplication-free CLUT-based decision-feedback equalizer (CLUT-DFE) and CLUT-based log-maximum a posteriori estimation with a fixed number of surviving states (CLUT-FS-MAP) to effectively compensate for channel distortions. Using the k-means clustering algorithm, the proposed CLUT significantly shortens the table size by reducing the number of independent variable values input. The table size of traditional LUTs increases exponentially with the filter tap number, making it impossible for LUTs to replace filters with tap numbers greater than 20. The CLUT breaks this limitation. The proposed schemes are experimentally performed in a C-band 61-Gb/s IM/DD optical on-off keying (OOK) system featuring a 3.65-GHz 3-dB bandwidth over a 100-km standard single mode fiber (SSMF). Compared to traditional LUT, CLUT reduces table size by about 10 orders of magnitude with the same performance. To the best of our knowledge, this study represents the first deployment of ultrasmall-size CLUT in equalization.