We demonstrate a thin-film lithium niobate polarizer with a resonant metal cavity for enhanced leakage. It achieves 42 dB polarization extinction ratio, 0.7 dB insertion loss, and crosstalk below -40 dB over a 160-nm bandwidth.

We combine the Volterra-series-inspired MoE mitigation framework with physics-inspired complex-valued CNN for fiber nonlinearities. Simulation results demonstrate 0.26 dB and 0.21 dB Q-factor gain over DBP of 10-step-per-span in 32 GBaud and 64 GBaud systems.

A PPIC-based OADC with an automatic defect-tolerant configuration algorithm is proposed. OADCs with reconfigurable resolutions have been demonstrated to achieve ENOB of 3/3.98/4.7-bit under realistic condition. Robustness analysis is also performed.

We propose and demonstrate a performance-enhanced optical quantizer by inverse design.An adjoint shape cooptimization method is used to optimize the boundaries of the optical quantizer,aiming to reduce the insertion loss(IL),improve the uniformity,and increase the bandwidth of the effective number of bits (ENOB).Meanwhile,the optimized shape maintains its deep ultraviolet (DUV) photolithography fabrication capability.We fabricate the device on a commercial silicon-on-insulator (SOI) platform.Measurement results show that the IL is reduced from 0.85 to 0.35 d B,and the uniformity is optimized from 1.21 to 0.24 d B at 1550 nm.The maximum ENOB increases to 3.31 bit,which is very close to the ideal value of 3.32 bit,and the bandwidth of the ENOB>3 bit is expanded to more than 50 nm.

A low-complexity adjusted logarithmic step-size (ALSS) iterative algorithm for nonlinear compensation in high-speed, high-order modulation WDM optical fiber communication systems is proposed. By adaptively optimizing the cumulative global step size, the ALSS algorithm effectively mitigates fiber nonlinearities, extending transmission distance and enhancing system robustness with significantly reduced computational overhead. To evaluate its performance, a 3-channel simulation system is built for 85 GBaud PDM-16QAM over 1600 km and 85 GBaud PDM-64QAM over 480 km, along with a single-channel experimental setup for 64 GBaud PDM-16QAM over 1213.56 km. In the 85 GBaud PDM-16QAM system, ALSS-10 StPS achieves Q-factor improvements of 0.60 dB and 0.92 dB over DBP-10 StPS and linear compensation, respectively, with a transmission reach extended by approximately 10% under the 7% FEC threshold. In case of the 85 GBaud PDM-64QAM transmission system, a 0.51 dB and 1.06 dB gain is observed over DBP-10 StPS and linear compensation at 4 dBm launch power. ALSS consistently outperforms LPF-DBP with only 54% of the complexity at all launch power. Experimental validation further confirms the practical feasibility of ALSS. Notably, ALSS-4 StPS surpasses LPF-DBP-4 StPS in Q-factor with only 54% complexity, and ALSS-2 StPS matches the performance of DBP-4 StPS with only 50% complexity.

A tunable Raised Cosine (tRC) LPF scheme is proposed for nonlinear compensation in optical systems. Experiments show a 0.31 dB Q-factor gain and 50% complexity reduction compared to Gaussian-LPF DBP.

We propose a precision-enhanced ADC scheme combining optical phase quantization and electrical amplitude quantization. Experimental results show a 2-bit improvement in ENOB compared to an unenhanced ADC at a sampling rate of 10Gs/s.

We demonstrate an integrated core-pumped 4-core erbium-doped fiber amplifier (4C-EDFA) that achieves a record-low differential core gain of 0.5 dB across the whole C-band. This is enabled by utilizing a 4C-EDF with a minimal core-dependent absorption coefficient and passive devices with low core-dependent loss. The 4C-EDFA also exhibits an average gain of 15.50 dB, an average output power of 22.5 dBm, and a maximum noise figure of 4.91 dB. Furthermore, simulations on a 4-core fiber (4CF) transmission link confirm that the proposed 4C-EDFA can support transmission exceeding 10,000 km with a minimal inter-core Q2 difference of only 0.5 dB. Here, Q2 is defined as the ratio of the mean received signal levels to the corresponding noise variances. It is a critical metric in optical systems to quantify the signal quality, which highlights its potential for high-capacity and long-haul uncoupled 4CF systems.

In this paper, a photonic dual-band radar receiving and processing technique based on a Stepped Linear Frequency Modulation (SLFM) signal has been proposed, achieving high-resolution radar detection driven by a low-rate signal source. The generation of dual-band radar signal with GHz-level bandwidth from SLFM sub-pulse signals with MHz-level bandwidth is utilized, significantly conserving the bandwidth resources of the radar driving signal source. Subsequently, a photonics-assisted dual-channel radar de-chirper cascaded with two electro-optic modulators is employed to realize wideband radar pulse compression processing covering both the X-band and Ku-band. Finally, the coherent fusion processing algorithm for dual-band radar is utilized, not only accomplishing an equivalent high-resolution radar ranging and Inverse Synthetic Aperture Radar (ISAR) imaging but also facilitating an anti-jamming radar detection. In simulation experiments, this paper has achieved the reception of dual-band SLFM radar signals with a sub-pulse number of 100 and a frequency coverage range of 8GHz-16GHz. Due to the coherent fusion processing algorithm, the radar target ranging and ISAR imaging with a resolution of similar to 2cm (equivalent bandwidth of 8GHz) successfully have been achieved.

We demonstrate a performance-enhanced optical quantizer by utilizing a shape optimization method. Simulation results show that the insertion loss is optimized from 0.22 dB to 0.1 dB, and the uniformity is optimized from 0.52 dB to 0.1 dB at 1550 nm. The effective number of bits (ENOB) is more than 3.2 bit over a large bandwidth of 80 nm (1518-1598 nm).

Fast-than-Nyquist (FTN) signaling improves spectrum efficiency (SE) by compressing the symbol interval and is a promising communication scheme. However, high compression rate means serious inter-symbol interference (ISI), so researchers have been exploring various low-complexity and efficient equalization methods. Delayed feedback sequence estimation (DDFSE) achieves a good tradeoff between complexity and equalization effect, and can achieve better equalization effect with a prefilter, but no one has compared which prefilter has the best BER performance for FTN signaling. We proposed a new prefilter based on unbiased maximum signal-to-interference-to-noise ratio (U-SINR) prefilter and mean square whitened matched filter (MS-WMF), and compared it with three other prefilters to find the best prefilter for medium and low time-squeezing factors. The simulation results show that in the vast majority of cases, we recommend MS-WMF, but new prefilter can be used to obtain better equalization effects when the signal-to-noise ratio is high and the time-squeezing factors is low.

The lossless path-averaged (LPA) approximation method is combined with the enhanced Gaussian noise (EGN) model, consequently inducing simple general closed-form expression for swift precise nonlinear interference estimation in optical transmission system with multi-arbitrary spans.

Nonlinear frequency division multiplexing (NFDM) is a promising optical transmission system that avoids fiber nonlinear effects, but suffering from unsatisfied spectral efficiency due to the guard interval (GI) devoting to combat the cumulative chromatic dispersion (CD), whose contribution to efficiency declination is quadratic to the bandwidth. In this paper, we propose what we believe to be a novel digital sub-band multiplexed (DSBM) NFDM system that can diminish the minimal required guard interval and therefore realize considerable information rate without reliability degradation. Through numerical simulation, we show that root raised cosine (RRC) shaped subcarriers are more suitable than the commonly used orthogonal frequency division multiplexing (OFDM) shaped ones in our DSBM-NFDM system. By multiplexing 15 sub-bands, we achieve a 129.7% improvement in effective information rate (EIR) compared to the single channel (SC) NFDM system with the same effective bandwidth. Meanwhile, we achieve a net information rate of 531.0 Gbps NFDM transmission over 960 km with SE of 3.86 bit/s/Hz, which is below the hard decision forward error correction (HD-FEC) threshold. This work is the first to introduce the DSBM technique into NFDM system, and it can be considered as a novel subcarrier shaping scheme for NFDM that has potential to advance its practical implementation.

The advanced logarithmic step size distribution (ALSS) compensation technique is proposed . To achieve similar performance, the computational complexity of the proposed algorithm is about 40% of the GLSS distribution algorithm and DBP algorithms.

Based on the distribution pattern of the noise, a scheme with joint modulation formats is proposed to resist noise in the b-modulated dual-polarization continuous spectrum NFDM system. Results show that the joint modulation formats can effectively improve the system performance.

We propose a novel 1-bit enhancement scheme for an MMI-based optical quantization, by introducing a high-level bit through a 1 x 3 phase modulation structure and a high-level quantizer. The scheme has low complexity because it does not introduce additional modulators or multiple optical channels; it simply adds one channel. We simulated the optical chip designed for this scheme on the lithium niobate on insulator (LNOI) platform and obtained results for insertion loss (IL) and effective number of bits (ENOB). The simulation results indicate that the chip can achieve >4bits in a wavelength range of 1500nm to 1590nm. The minimum loss of the chip is 0.77dB at 1544.5nm. And the ENOB at 1544.5nm reaches 4.30bits.