Inter-satellite optical wireless communication (Is-OWC) demands compact, power-efficient transmitters for high-capacity, long-distance links. We focus on parallel amplification of multimode Laguerre-Gaussian (LG) beams using a few-mode erbium-doped fiber amplifier (FM-EDFA). A corresponding theoretical model with space radiation-induced attenuation (RIA) is established to investigate LG-mode gain and fidelity. The analysis shows that the LG fidelity is determined by differential modal gain (DMG) and time-delay difference between linearly polarized (LP)-mode components. Under circularly symmetric LP01-mode pumping, the FM-EDFA possesses a high LG(01) mode fidelity of >93% after phase compensation. By optimizing the pump combination of LP01, LP11e and LP11o modes, four LG modes (LG(00), LG(01), LG(02), LG(10)) can be simultaneously amplified to >26 dB, with a gain difference of <3 dB. In the presence of space radiation, the LG-mode gain decreases with an average slope of about -0.761 dB/krad. The fidelities of the LG(00), LG(01), and LG(02) modes are almost fixed, while that of the LG(10) mode increases with radiation dose. In addition, we present three phase-compensation strategies for high fidelity, those of using phase shifters, in-fiber long-period fiber grating, and spatial light modulator. In a word, multiple LG modes can be simultaneously amplified by using the currently available LP-mode EDFAs with automatic gain control and phase compensation, capable of satisfying the Is-OWC's requirements for compactness, high gain and radiation-awareness.

Chaos-based optical fiber communication technology offers strong protection for the security of modern optical information transmission. However, chaotic systems are susceptible to dynamic degradation in communication scenarios with finite-precision computation. This paper investigates the impact of dynamic degradation on the performance of secure optical fiber communication systems. To address this issue, we propose an encryption framework based on a cellular automaton based discrete chaotic system (CA-DCS). The CA-DCS employs an 8-state cellular automaton as the chaotic source, generating a highly random discrete sequence through finite-state chaotic evolution. Chaotic confusion and constellation masking are then applied to mask the physical layer data. We demonstrate encrypted data transmission at 32 GBaud over 400 km standard single-mode fiber. The results indicate that dynamic degradation undermines the physical layer security masking and exacerbates transmission impairments, resulting in a performance penalty of 5.3 dB. The proposed scheme maintains robust chaotic masking even under finite-precision computation and offers an expandable key space of 10240. This study highlights key performance-limiting factors in chaos-based optical fiber communication and suggests a promising direction for future research in secure chaotic communication technologies.

Stimulated Raman scattering (SRS)-induced inter-channel power transfer introduces global cross-channel coupling in the amplitude domain, causing power fluctuations across the spectrum that further interact with accumulated Kerr nonlinearities. The C + L-band system modeling and generalization become challenging under various nonlinear effects, especially with power pre-tilt. In this Letter, we propose an enhanced self-attention-assisted multi-channel waveform modeling to achieve efficient and accurate modeling with SRS and non-flat launch power. To capture these cross-channel and long-range dependencies, we apply rotary positional encoding to the query (Q) and key (K) matrices in the attention mechanism. Benefitting from enhanced self-attention, we realize waveform modeling with strong generalization ability across different non-flat launch power profiles and transmission distances in ultra-wideband (UWB) wavelength-division multiplexing (WDM) systems. We compare the split-step Fourier method (SSFM) with the proposed method over a 10-span link at the optimal launch power, and the Q-factor differs from SSFM by only 0.31 dB. In a 5-span scenario with pronounced nonlinear effects, our method reduces runtime by 99.4% while maintaining a Q-factor deviation of just 0.24 dB from SSFM.

We designed an OFEC for coherent satellite communication and developed a 2.894 Gbps FPGA-based system (15.3% overhead). Experimentally, the system achieved a high-sensitivity of - 57.6 dBm with a 2 × 10 − 2 BER decoding threshold.

We numerically demonstrate a multi-channel wideband secure key distribution scheme using commonly driven Fabry-Perot lasers. A phase-modulation and dispersion loop enables independent channel extraction, achieving ~14.6 Gbps per-channel key rates with low key inconsistency.

We propose a digital-twin-assisted forward-inverse dual-network framework that achieves accurate pump-failure detection and performs online re-optimization within 2.5 s, restoring mean GSNR to the pre-fault level in C+L-band networks.

Introducing adjacent S-band to C +L-band has been an effective and feasible solution to significantly increase the transmission capacity of single fiber. However, stimulated Raman scattering (SRS) leads to serious performance degradation of S-band, which in turn limits the overall capacity improvement and performance uniformity. In this work, we propose a two-step optimization method for Raman amplifier design and launch power optimization to maximize the transmission capacity, meantime guaranteeing the general signal to noise ratio (GSNR) uniformity and improving the pump energy efficiency. We employ a partially parameter-fitted ISRS-GN model to reduce quality of transmission (QoT) computation time and accelerate the search process through a particle swarm optimization (PSO) implemented with Python's multiprocessing module. We present an 18 THz S + C + Lband transmission system that achieves more than four times the capacity of conventional C-band, while maintaining a full-band GSNR flatness within +/- 0.45 dB. This is achieved using three Raman pumps with a total power budget of 668 mW. The hybrid amplification of the three Raman pumps demonstrates the optimal energy efficiency. This method can be used for designing arbitrary optical fiber UWB WDM systems before practical testing.

Bipolar pulse amplitude modulation (BPAM) can improve power efficiency by leveraging the bipolar optical field. However, the maximum likelihood (ML) detection for BPAM grows exponentially in complexity as the modulation order and channel memory length increase. In this paper, we propose what we believe to be a novel multi-stage detector for BPAM to eliminate multi-level impairments while maintaining tolerable complexity. A key innovation of the multi-stage detector is the extension of sign estimation to general shaping pulses using the Viterbi algorithm. The detector effectively decouples and compensates for the signal impairments from the transmitter, the polarity loss of square-law detection, and the receiver in a sequential multi-stage process. In a 1-km C-band transmission experiment using 50-Gbaud BPAM-8 signals, the proposed detector achieves a bit-error ratio (BER) below 2.2 × 10-2 (the 20% soft-decision forward error correction (SD-FEC) threshold) across a range of roll-off factors. As a comparison, conventional ML detection fails to recover the signal due to the impractical computational complexity. These results demonstrate the effectiveness of our multi-stage detector in addressing current limitations.

With the development of the internet, the security of the optical data center interconnections (DCIs) has become a highly concerned topic. However, current security schemes face limitations such as the absence of viable key distribution mechanisms and the sensitivity to channel condition fluctuations. To address the above challenges, we propose a novel physical layer encryption scheme for WDM-based DCIs that combines encryption with channel conditions. Considering the need for dynamic key updates in DCIs, we first propose embedding plaintext-related keys (PRKs) into wavelength-related pilot (WP) positions for distribution. Leveraging their controllable size to combat impairments under varying channel conditions enables dynamic error propagation. Additionally, the proposed recombination mechanism supplements DNA encryption, allowing data transmission across different channels during inter-strand base exchange. The two channel 111.12 Gb/s discrete multi-tone (DMT) signal transmissions are successfully demonstrated. These results indicate that this scheme can transmit data on different paths. The key space can reach 10360 and the complexity is about O(n). Furthermore, the key can be transmitted without redundancy and strongly relies on channel conditions to control security. Moreover, this scheme can alleviate the damage caused by high peak to average power ratio (PAPR), with an increase of approximately 0.5dBm at the forward error correction (FEC) limit.

High-speed digital subcarrier multiplexing (DSCM) coherent optical system is more sensitive to practical impairments in electrical and optical components, particularly the transmitter (Tx) in-phase/quadrature (IQ) impairments, including IQ amplitude imbalance, phase imbalance, and timing skew. In this paper, we propose a joint estimation based on paired subcarriers (JE-PS) with a designed training sequence for Tx IQ impairments. The training sequence on one pair of symmetric subcarriers is designed to be negative conjugate, and another pair of symmetric subcarriers is conjugate. Then we can use two subcarriers located in the negative or positive frequency region to extract the Tx IQ amplitude imbalance and IQ skew in the frequency domain. Since the IQ skew and subcarrier frequency induce extra phase interference, we use negative and positive paired subcarriers to eliminate the interference and calculate the IQ phase imbalance. Theoretical analysis and simulation results indicate that JE-PS methods are insensitive to the modulation format and laser phase noise. Besides, it is also robust to the transceiver IQ impairments and ASE noise. Finally, the JE-PS methods are experimentally verified in a 50-Gbaud four-subcarrier PM-16QAM DSCM signal transmission system. The estimation accuracy of Tx IQ amplitude, phase imbalance, or skew is within 0.25 dB, 1°, or 0.2ps, respectively.

Optical frequency division (OFD) converts ultrastable optical frequencies to microwaves via an optical frequency comb, generating microwave oscillators with record-low phase noise and time jitter. However, conventional OFD systems face a notable trade-off between division complexity and noise suppression because of severe thermal and technical noise in optical references. Here, we address this challenge by using common-cavity bicolor Brillouin lasers as references, operating at the fundamental quantum noise limit with a 10-microhertz Schawlow-Townes linewidth. Enabled by these ultracoherent lasers, our OFD system uses a markedly simplified comb divider with an unprecedented division factor of 10, producing a 10-gigahertz microwave signal with exceptional phase noise of -65 decibels relative to the carrier per hertz at 1-hertz offset, -155 decibels relative to the carrier per hertz at 10-kilohertz offset, and -172 decibels relative to the carrier per hertz at 10-megahertz offset. Leveraging this purity, we implement broadband synthesis from 5 to 20 gigahertz with millisecond tuning. This work redefines the trade-off between noise suppression and division complexity in OFD, paving the way for compact, high-performance microwave synthesis for next-generation atomic clocks, quantum sensors, and low-noise radar systems.

We propose a joint estimation of transmitter IQ amplitude, phase imbalance and time skew based on paired subcarriers using a designed training sequence, with the measurement error within 0.2dB, 1°, and 0.2ps, respectively.

The growing data exchange among consumer-electronic (CE) devices demands encryption methods that balance security, efficiency, and real-time performance. Thus, we introduce a four-dimensional Dynamic Feedback Memristive Chaotic System (4D-DFMCS) and then propose a chaotic encryption method based on block-wise diffusion. Specifically, a 4D-DFMCS with enhanced chaotic properties is constructed by introducing the nonlinear properties of the memristor. It has a larger chaotic range and can improve chaotic dynamic degradation compared with traditional systems, thereby enhancing the security and stability of the encryption algorithm. Using this system for block data diffusion, a novel block-wise random diffusion mechanism is designed. By introducing global dependencies, this method significantly reduces the risk of system deciphering due to eavesdropping attacks. This further enhances the protection of sensitive visual data on CE devices against diverse attacks. In addition, the pre-sharing defects of initial value are prevalent in existing chaotic cryptographic systems, so a novel initial key generation and dynamic update mechanism based on Elliptic Curve Diffie-Hellman Ephemeral (ECDHE) protocol is developed. Finally, the result shows that the key space reaches 4.26 & times;10(178) . The Number of Pixels Change Rate (NPCR) and Unified Average Changing Intensity (UACI) values are obtained as 99.6056% and 33.4333%. It further demonstrates the scheme's robustness against eavesdropping under SNR constraints, with decrypted outputs appearing noise-like below 15 dB. These results demonstrate the strong security performance of the proposed scheme in CE applications requiring both high confidentiality and real-time processing.

Optical chaotic synchronization between coupled nonlinear lasers underpins most chaos-based applications, including complex laser network dynamics, secure communication, key distribution, and reinforcement learning. In free-space links, however, chaotic synchronization is highly vulnerable to stochastic fluctuation induced by atmospheric turbulence, which results in temporal injection imbalances at symmetric receivers and triggers intermittent desynchronization. Here, we introduce a full Poincaré vector beam-enabled balanced-injection synchronization (BIS) mechanism, which passively mitigates coupling fluctuations and preserves injection symmetry through a complementary metasurface pair, without requiring any channel estimation or active control. Over a 3.2 km urban link under moderately strong turbulence, BIS suppresses coupling power fluctuations by a factor of 4.6 (from 0.4511 to 0.0975). It eliminates desynchronization events and increases the high-quality synchronization probability from 58.6% to 91.0%. This enables a record-high bit rate-distance product of 720 Gbps \cdot km, reducing communication interruption probability by up to 77% compared to Gaussian beam transmission. Our innovative strategy bridges the gap between nanophotonics and engineering optics, offering a new insight into advancing next-generation LiDAR, secure communication, and integrated sensing and communication systems in turbulent environments.

Fast and low-cost monitoring of coherent transmitter is achieved using a 3-GHz photodetector and an alternating iterative algorithm. It enables accurate estimation of IQ skew (0.16 ps) and frequency response within only 10 iterations.

Joint estimation of optical signal-to-noise ratio (OSNR) and mode-coupling induced crosstalk (XT) is essential for the management of mode-division multiplexing (MDM) networks. We propose a multi-task joint optical performance monitoring (OPM) scheme based on a Self-Attention Long Short-Term Memory (SA-LSTM) network. By utilizing dual-domain statistical feature spectrum amplitude histograms (SAH) and filtered waveform amplitude histograms (FWAH), the proposed method characterizes frequency and temporal impairment information without requiring complex digital signal processing (DSP). Within the proposed architecture, the self-attention mechanism captures global cross-bin dependencies, while the LSTM network models the sequential evolution of the histogram profiles. Simulation results based on a 3-mode 25-GBaud QPSK MDM system demonstrate that the SA-LSTM framework effectively decouples amplified spontaneous emission (ASE) noise and modal crosstalk. The model achieves 100% accuracy within a 0.5 dB error margin (Acc@0.5 dB) and a root mean square error (RMSE) of 0.0798 dB for OSNR estimation (5 to 20 dB), and a 99.12% Acc@0.5 dB with a 0.1276 dB RMSE for XT estimation (-10 to -3 dB). The proposed SA-LSTM scheme provides a robust and high-accuracy OPM solution for dynamic MDM optical networks.

Ultrastable lasers (USLs), characterized by exceptional phase coherence and frequency stability, are indispensable for various fields including fundamental physics, timekeeping, spectroscopy, metrology, quantum computing, and astronomy. However, conventional USLs are costly and bulky, limiting their availability to specialized laboratories. Here, we introduce a new concept for the multiplication of USLs using a microcavity Kerr frequency comb on-chip. We demonstrate that the Kerr microcomb can be all-optically locked to both the optical reference and microwave reference, allowing the microcomb teeth to inherit the spectral purity of the frequency standards and become an array of USLs. In our experiment, 100 Kerr microcomb teeth achieve integral linewidth significantly below 1 Hz and fractional frequency instability on the order of over 1-second average time, making them 100 state-of-the-art USLs. Our method holds potential to proliferate the use of ultrastable lasers in a wide range of applications.

The mode extinction ratio (ER) of a mode multiplexer/demultiplexer (M-MUX/DEMUX) has a significant influence on the mode crosstalk performance for mode division multiplexing (MDM) optical transmission systems and multi-dimensional optical switching nodes. A mode interference demultiplexing (MIDM) scheme based on the Mach-Zehnder interferometer (MZI) structure is put forward to further improve the mode demultiplexing performance. Its operating principle is based on the coherent properties between mode crosstalk and the original signal introduced by the M-DEMUX. Taking the two-mode demultiplexer (2M-DEMUX) composed of LP01 and LP11 mode selective couplers (MSCs) as an example, we illustrate the advantages of the MIDM scheme in ER equalization and improvement, and the involved interference unit is designed on the basis of the silicon-on-insulator (SOI) platform. The simulation shows that the balanced ER is up to 28.34 dB, identical with the theoretical results. The application of the MIDM scheme to the six-mode DEMUX is also discussed, with an ER improvement of 2.57 dB for the worst LP21 mode.