AnAdapter module is developed to increase the generalization capability of the end-to-end learning system for optical fiber communication systems. The Adapter has an interpretable structure and can be inserted and used without changing the original signal processing structure. The Adapter module can improve the system generalization capability and achieve the correct demapping within the transmission distance fluctuation range of $\pm$100 km and the power fluctuation range of $\pm$0.5 dBm. In addition, using an Adapter can improve the performance of optical communication by modifying the equalization algorithm without altering the structure of the existing transmission system.

This paper proposes an autoencoder (AE)-based probabilistic shaping (PS) framework for coherent optical fiber systems that, for the first time, explicitly incorporates equalization-enhanced phase noise (EEPN). By modeling EEPN with both a simplified analytical channel and a more accurate physical mechanism, we train AE-based PS distributions to maximize generalized mutual information (GMI). Our results reveal that the computationally efficient analytical model yields shaping gains nearly on par with its physically derived counterpart. Furthermore, PS distributions learned under linear assumptions retain robust performance in nonlinear fiber channels, particularly at higher laser linewidths, underscoring strong resilience to model mismatch. We also examine the adaptability of these AE-trained constellations across various system parameters, including linewidth and launch power, providing insights into their practical deployment. Overall, this work highlights the feasibility of EEPN-aware PS and demonstrates the effectiveness of simplified channel models for reducing design complexity, offering guidelines for implementing AE-based PS in next-generation coherent optical communication systems.

This paper presents a comprehensive review of machine learning (ML) in optical fiber communications, particularly in channel modeling. It discusses the evolution from conventional methods to ML-based approaches that aim to enhance predictive and computational efficiency. Specifically, the discussions categorize ML methodologies into data-driven and principle-driven approaches. The former treats channel modeling as a “black box” providing rapid modeling capabilities at the expense of transparency and substantial data requirements. In contrast, the latter integrate physical principles into the ML-based system, enhancing model interpretability and reducing data dependency. In addition, the emergence of hybrid models that combine the strengths of both approaches is explored. This classification provides a structured overview of how ML is reshaping channel modeling in optical fiber communications, underscoring its potential to improve system design and exploring advanced nonlinear dynamics in optical fiber communication systems.

Transparent wood (TW) is a sustainable composite material with high optical transmittance and excellent mechanical properties. Nanoparticles, dyes and quantum dots can be added in a controlled manner for new functionalities relying on the light scattering properties of the composite. The scattering properties of 3D TW models of cellular microstructure are investigated numerically using geometrical optics. A group of 3D TW material models with controlled microstructural parameters are generated based on an analytical method. A ray tracing approach is adopted to model scattering in these complex materials. Effects from different material parameters on ray scattering are analyzed. A virtual camera or virtual eye to render images positioned behind a TW plate is simulated using backward ray tracing. The blurred impression in human eyes of real objects viewed through a TW “window” can then be mimicked.

Development of Data Center based computing technology require energy efficient high-speed transmission links. This leads to optical amplification-free intensity modulation and direct detection (IM/DD) systems with low complexity equalization compliant with IEEE standardized electrical interfaces. Switching from on-off keying to multi-level pulse amplitude modulation would allow to reduce lane count for next generation Ethernet interfaces. We characterize 106.25 Gbaud on-off keying, 4-level and 6-level pulse amplitude modulation links using two integrated transmitters: O-band directly modulated laser and C-band externally modulated laser. Simple feed forward or decision feedback equalizer is used. We demonstrate 106.25 Gbaud on-off keying links operating without forward error correction for both transmitters. We also show 106.25 Gbaud 4-level and 6-level pulse amplitude modulation links with performance below 6.25% overhead hard-decision forward error threshold of 4.5 × 10^-3. Furthermore, for EML-based transmitter we achieve 106.25 Gbaud 4-level pulse amplitude modulation performance below KP-FEC threshold of 2.2 × 10^-4. That shows that we can use optics to support (2x)100 Gbps Ethernet on single lambda at expense of simple forward error correction.

This chapter is an integrated overview of organic solid-state tunable laser oscillators that adds an additional and complimentary perspective to the material covered in chapters 5 to 9. Emphases are on miniaturized tunable organic lasers using DFB, VECSOL, and waveguide architectures. Furthermore, the use of dye-doped transparent wood, as an all-organic solid-state gain medium, is described and discussed.

The significant influence of equalization enhanced phase noise on the performance of long-haul EDFA-amplified optical fiber communication systems has been investigated in this paper. A 128-Gbaud DP-16QAM multi-channel 2400 km Nyquist-spaced optical transmission system has been considered, with the application of the electronic dispersion compensation and the digital nonlinearity compensation schemes.

We experimentally demonstrate and compare EML- and DML-based optical interconnects with 106.25 Gbaud NRZ-OOK and PAM4 for computing applications. The results show that both transmitters can be used to enable optical -amplification-free transmissions with low-complexity DSP.

We experimentally demonstrate an O-band single-lane 200 Gb/s intensity modulation direct detection (IM/DD) transmission system using a low-chirp, broadband, and high-power directly modulated laser (DML). The employed laser is an isolator-free packaged module with over 65-GHz modulation bandwidth enabled by a distributed feedback plus passive waveguide reflection (DFB+R) design. We transmit high baud rate signals over 20-km standard single-mode fiber (SSMF) without using any optical amplifiers and demodulate them with reasonably low-complexity digital equalizers. We generate and detect up to 170 Gbaud non-return-to-zero on-off-keying (NRZ-OOK), 112 Gbaud 4-level pulse amplitude modulation (PAM4), and 100 Gbaud PAM6 in the optical back-to-back configuration. After transmission over the 20-km optical-amplifier-free SSMF link, up to 150 Gbaud NRZ-OOK, 106 Gbaud PAM4, and 80 Gbaud PAM6 signals are successfully received and demodulated, achieving bit error rate (BER) performance below the 6.25%-overhead hard-decision (HD) forward-error-correction code (FEC) limit. The demonstrated results show the possibility of meeting the strict requirements towards the development of 200 Gb/s/lane IM/DD technologies, targeting 800 Gb/s and 1.6 Tb/s LR applications.

We experimentally demonstrate a Long-Wave IR FSO link with a 9.15-μm directly modulated quantum cascade laser at room temperature. Up to 8.1 Gb/s PAM8 transmission over 1.4 meter is achieved with a wideband MCT detector.