Guoxiang Xu
Research Subjects
Guoxiang Xu has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
Rendong Xu
Jiangsu Key Laboratory for Subsea Communications and Sensing, Changshu, China
4 shared publications- Gangxiang Shen
Soochow University
3 shared publications - Lin Sun
Suzhou Key Laboratory of Advanced Optical Communication Network Technology, School of Electronic and Information Engineering, Jiangsu Engineering Research Center of Novel Optical Fiber Technology and Communication Network, Soochow University, Suzhou, China
3 shared publications - Gordon Ning Liu
Soochow University
2 shared publications - Haocai Huang
Zhejiang University
2 shared publications Wei Yin
Hengtong Huahai Technology Company Ltd., Changshu, China
2 shared publications
Latest Publications
- Open Access
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Journal
- Vol. 18 (2) , 1-6
Multidimensional Bit-Wise Sparse Decoding for Residual IC-XT Mitigation in Optical Long-Haul CC-MCF Links
Coupled-core multicore fiber (CC-MCF) transmissions rely on sufficient taps of multiple-input and multiple-output (MIMO) equalizers which should cover the channel temporal pulse response raised by the elevated spatial mode dispersion of fibers. Otherwise, the residual inter-core crosstalk (IC-XT) after MIMO will impede the signaling quality when employing finite taps of MIMO equalizers. In this study, we experimentally examine how residual IC-XT evolves by evaluating the covariance among MIMO-equalized spatial tributaries in multiple-span CC-MCF transmission links as the MIMO equalizers' tap length decreases. As indicated by results, increasing covariance values happen along with the stronger correlations among spatial modes after MIMO equalization with using shorter-tap equalizers. Multidimensional bit-wise decoding (MD-BWD) can process the spatial domain-constructed multidimensional symbols, offering adaptive decoding capability for the strongly-correlated spatial modes but introducing significant DSP computational complexity. To address this, we propose a sparse representation of log-likelihood ratio calculation in MD-BWD to minimize the reference labels needed for soft information acquisition during decoding procedure. Experimental results for optical 20-Gbaud DP-QPSK transmissions over 1500-km CC-4CFs show that the proposed multidimensional sparse decoding method reduces complexity by 96.5% compared to traditional MD-BWD, while maintaining Q factors below the 20% overhead FEC threshold using short-tap MIMO equalizers. This advancement in reducing DSP complexity could facilitate the practical implementation of MD-SD signaling to mitigate residual IC-XT issues.
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Journal
- Vol. 18 (2) , 1-6
Open AccessMultidimensional Bit-Wise Sparse Decoding for Residual IC-XT Mitigation in Optical Long-Haul CC-MCF Links
Coupled-core multicore fiber (CC-MCF) transmissions rely on sufficient taps of multiple-input and multiple-output (MIMO) equalizers which should cover the channel temporal pulse response raised by the elevated spatial mode dispersion of fibers. Otherwise, the residual inter-core crosstalk (IC-XT) after MIMO will impede the signaling quality when employing finite taps of MIMO equalizers. In this study, we experimentally examine how residual IC-XT evolves by evaluating the covariance among MIMO-equalized spatial tributaries in multiple-span CC-MCF transmission links as the MIMO equalizers' tap length decreases. As indicated by results, increasing covariance values happen along with the stronger correlations among spatial modes after MIMO equalization with using shorter-tap equalizers. Multidimensional bit-wise decoding (MD-BWD) can process the spatial domain-constructed multidimensional symbols, offering adaptive decoding capability for the strongly-correlated spatial modes but introducing significant DSP computational complexity. To address this, we propose a sparse representation of log-likelihood ratio calculation in MD-BWD to minimize the reference labels needed for soft information acquisition during decoding procedure. Experimental results for optical 20-Gbaud DP-QPSK transmissions over 1500-km CC-4CFs show that the proposed multidimensional sparse decoding method reduces complexity by 96.5% compared to traditional MD-BWD, while maintaining Q factors below the 20% overhead FEC threshold using short-tap MIMO equalizers. This advancement in reducing DSP complexity could facilitate the practical implementation of MD-SD signaling to mitigate residual IC-XT issues.
- 7 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Emerging and Selected Topics in Power Electronics
- Vol. 14 (1) , 258-268
A DC Current-Source Input and Voltage-Source Output Converter for Wide Range Variation of Load Power
This paper proposed a current-source input and voltage source output converter for wide range variation of load power. The challenge of constant current conversion is the input voltage tends to diverge to the highest at reduced load condition and tends to lowest at increased load condition, as the current cannot be varied abruptly. The proposed converter features regulated input characteristics and output voltage from rated power to no-load, provides a current-freewheel branch composed of MOSFET and transformer for the constant input current, which solves the problem of input voltage disperse to the highest; besides, the energy stored in the floating capacitor is used as transient energy for the load power increase, which avoids the input voltage disperse to 0V as it changes from no-load to full-load, and the energy of the floating capacitor is derived from the transformer magnetic reset. Furthermore, the typically constant current supply systems operate at low currents and the power is limited by the input voltage, so the proposed converter employs two capacitors charging in series and discharging interleaved, which improved the input voltage level and reduced the output filter inductance. Moreover, the lagged freewheel switches realizes zero-voltage turn-on, which is helpful for efficiency. A prototype with 2A dc input and 250V/530W output is built to verify the feasibility of the proposed topology.
- 7 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Emerging and Selected Topics in Power Electronics
- Vol. 14 (1) , 258-268
A DC Current-Source Input and Voltage-Source Output Converter for Wide Range Variation of Load Power
This paper proposed a current-source input and voltage source output converter for wide range variation of load power. The challenge of constant current conversion is the input voltage tends to diverge to the highest at reduced load condition and tends to lowest at increased load condition, as the current cannot be varied abruptly. The proposed converter features regulated input characteristics and output voltage from rated power to no-load, provides a current-freewheel branch composed of MOSFET and transformer for the constant input current, which solves the problem of input voltage disperse to the highest; besides, the energy stored in the floating capacitor is used as transient energy for the load power increase, which avoids the input voltage disperse to 0V as it changes from no-load to full-load, and the energy of the floating capacitor is derived from the transformer magnetic reset. Furthermore, the typically constant current supply systems operate at low currents and the power is limited by the input voltage, so the proposed converter employs two capacitors charging in series and discharging interleaved, which improved the input voltage level and reduced the output filter inductance. Moreover, the lagged freewheel switches realizes zero-voltage turn-on, which is helpful for efficiency. A prototype with 2A dc input and 250V/530W output is built to verify the feasibility of the proposed topology.
- Open Access
- 28 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Technology Letters
- Vol. 37 (21) , 1249-1252
Advanced Integration of Sensing and Communication With High DSP Compatibility for SMART Network
Concept of the Scientific Monitoring and Reliable Telecommunications (SMART) network involves utilizing submarine communication cables to monitor the undersea environment. Traditionally, this integrated sensing and communication (ISAC) application has relied on separate wavelength bands for sensing information switching among all sensing joints and shores, necessitating additional hardware to exchange and demodulate the sensing information. In this study, we introduce a novel ISAC configuration tailored for the SMART network, with demonstrating the viability of detecting modulated sensing information using a communication-compatible digital signal processor (DSP). By applying frequency modulation (FM) and optical single-sideband (SSB) modulation to the sensing information obtained by the electronic sensors, we convert the information into multiple frequency subcarriers that reside near the communication spectrum without interference. We experimentally validated the proposed ISAC configuration using a two-span single-mode fiber (SMF) communication link, achieving real-time temperature monitoring on an FPGA platform at the sensing joint between two 50-km span SMFs. Furthermore, our findings indicate that employing high-order frequency subcarriers and an analog frequency multiplier for frequency chirp amplification of the FM sensing information enhances temperature estimation accuracy when using the communication-compatible DSP. With this approach, we achieved a temperature monitoring resolution of 0.0625°C, which matches the resolution limit of the DS18B20 sensing probe used, while maintaining full compatibility with 16-GBaud DP-QAM16 coherent communication data transmissions. At last, the scalability of the proposed ISAC configuration for the multiple-span SMART cable is further discussed.
- 28 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Technology Letters
- Vol. 37 (21) , 1249-1252
Open AccessAdvanced Integration of Sensing and Communication With High DSP Compatibility for SMART Network
Concept of the Scientific Monitoring and Reliable Telecommunications (SMART) network involves utilizing submarine communication cables to monitor the undersea environment. Traditionally, this integrated sensing and communication (ISAC) application has relied on separate wavelength bands for sensing information switching among all sensing joints and shores, necessitating additional hardware to exchange and demodulate the sensing information. In this study, we introduce a novel ISAC configuration tailored for the SMART network, with demonstrating the viability of detecting modulated sensing information using a communication-compatible digital signal processor (DSP). By applying frequency modulation (FM) and optical single-sideband (SSB) modulation to the sensing information obtained by the electronic sensors, we convert the information into multiple frequency subcarriers that reside near the communication spectrum without interference. We experimentally validated the proposed ISAC configuration using a two-span single-mode fiber (SMF) communication link, achieving real-time temperature monitoring on an FPGA platform at the sensing joint between two 50-km span SMFs. Furthermore, our findings indicate that employing high-order frequency subcarriers and an analog frequency multiplier for frequency chirp amplification of the FM sensing information enhances temperature estimation accuracy when using the communication-compatible DSP. With this approach, we achieved a temperature monitoring resolution of 0.0625°C, which matches the resolution limit of the DS18B20 sensing probe used, while maintaining full compatibility with 16-GBaud DP-QAM16 coherent communication data transmissions. At last, the scalability of the proposed ISAC configuration for the multiple-span SMART cable is further discussed.
- 22 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Technology Letters
- Vol. 37 (1) , 5-8
PMD Impact on MCF Submarine Cables: Reassessing Outage Performance of Transmissions
In this study, we designed and fabricated a cable accommodating 4-core and 7-core multicore fibers (MCFs). Fiber characteristics including attenuation, dispersion and polarization mode dispersion (PMD) during the cabling procedure were measured and analyzed. We observed a significant degradation in PMD for MCFs, particularly for those outer cores proximate to the cladding. Thus it requires longer taps of MIMO equalizers to sufficiently equalize at the receiver for reliable communications at the subsea distance. To assess the PMD impact on outage performance with limitations of MIMO equalizers’ taps, we developed a C-band 8000-km submarine transmission model which was calibrated with measured parameters during cabling, for qualifying 4- and 7-core MCFs in practical use. Simulations performed on the calibrated model could predict the outage performance for ensuring the communication reliability. More than 2-dB SNR margin is required to prevent the PMD-induced outage with using 300-taps MIMO equalizer at the receiver. Based on these comprehensive tests and simulations, we emphasize the importance of carefully managing PMD and ensuring adequate system SNR margins during the practical cabling and deployment of optical MCFs submarine cables.
- 22 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Technology Letters
- Vol. 37 (1) , 5-8
PMD Impact on MCF Submarine Cables: Reassessing Outage Performance of Transmissions
In this study, we designed and fabricated a cable accommodating 4-core and 7-core multicore fibers (MCFs). Fiber characteristics including attenuation, dispersion and polarization mode dispersion (PMD) during the cabling procedure were measured and analyzed. We observed a significant degradation in PMD for MCFs, particularly for those outer cores proximate to the cladding. Thus it requires longer taps of MIMO equalizers to sufficiently equalize at the receiver for reliable communications at the subsea distance. To assess the PMD impact on outage performance with limitations of MIMO equalizers’ taps, we developed a C-band 8000-km submarine transmission model which was calibrated with measured parameters during cabling, for qualifying 4- and 7-core MCFs in practical use. Simulations performed on the calibrated model could predict the outage performance for ensuring the communication reliability. More than 2-dB SNR margin is required to prevent the PMD-induced outage with using 300-taps MIMO equalizer at the receiver. Based on these comprehensive tests and simulations, we emphasize the importance of carefully managing PMD and ensuring adequate system SNR margins during the practical cabling and deployment of optical MCFs submarine cables.