With the growing demand for data traffic, new types of optical fiber have been proposed to address the capacity crisis of existing single-mode fiber infrastructure. Space-division multiplexed (SDM) systems promise to exploit multiple spatial channels for parallel data transmission. The aim of this study is to provide a systematic approach to experimentally investigate the fluctuations of random coupling in the temporal domain of coupled multi-core fibers (MCFs). Understanding these temporal dynamics is crucial to setting requirements for digital signal processing and to developing robust signal recovery algorithms. This is also the first step towards dynamic channel models of these fibers. One of the key requirements for untangling the signals after transmission is accurate knowledge of channel behavior in both the time and frequency domains. We present a statistical and experimental analysis of the temporal dynamics of random coupling in strongly coupled MCFs. Based on a statistical time-drift model, we have experimentally characterized the temporal response of coupled MCFs and their statistical properties, such as the temporal autocorrelation function and the probability density function of fractional optical power.

Dual-pump fiber optical parametric amplifiers offer wide, flat gain spectra across arbitrary wavelengths, often outperforming the use of single-pump designs. However, gain flatness depends on the pump frequencies’ center of gravity relative to the nonlinear fiber’s zero-dispersion wavelength. This study experimentally investigated the gain, gain flatness, and noise figure of degenerate dual-pump fiber-optic phase-sensitive amplification at various separations between these points. We demonstrated 38 dB gain with 1.5 dB gain flatness variation over a 30 nm bandwidth as well as a noise figure below the conventional 3 dB quantum limit for a degenerate signal.

We present PAM-2 and PAM-4 modulated high-speed 850 nm vertical-cavity surface-emitting laser (VCSEL) based interconnects operating across the temperature range -60°C to 140°C. Two different multiple quantum well VCSEL designs were used along with forward error correction, electronic pre-emphasis, and receiver-side equalization techniques. A three-tap feed-forward equalizer was implemented as a pre-emphasis filter, and a least-mean-square equalizer was used at the receiver. We demonstrate successful PAM-4 transmission at 70 Gb/s from -60°C to 100°C, 50 Gb/s at 125°C, and 42 Gb/s at 140°C. The FEC overhead used in the experiments is 3.1%.

We studied linear and nonlinear coupling in coupled multicore fiber (MCF) experimentally and numerically. We investigated wavelength and polarization-dependent coupling and effects of coupling strength on parametric interactions. Results show that nonlinear responses depend sensitively on the longitudinally varying coupling coefficient in real MCF. (c) 2025 The Author(s)

Optical phase-sensitive amplifiers (PSAs) leverage nonlinear interactions to achieve noise figures below the conventional quantum limit. Dual-pump phase-insensitive amplification has been studied in wavelength conversion, phase conjugation and spectral inversion, but phase-sensitive amplification remains challenging due to stimulated Brillouin scattering (SBS) in optical fibers. While pump-phase modulation helps mitigate SBS, it broadens the idler spectrum, complicating practical implementation. We demonstrate a counter-phase pump modulation approach to suppress idler spectral broadening using optical coherent combining and delayed self-heterodyne techniques to achieve excellent noise figure (NF) performance in dual-pump PSAs. Our methods achieve over 30 dB suppression of residual phase modulation tones and significantly improve the idler spectrum. We obtained a 20 dB gain with an NF of 1.45 dB and a minimal NF penalty of 0.2 dB due to pump-phase modulation. This achievement enables PSAs to support optical communication, wavelength conversion, and deep-space communication applications.

Four-wave mixing is a nonlinear optical phenomenon that can be used for wideband low-noise optical amplification and wavelength conversion. It has been extensively investigated for applications in communications 1 , computing 2 , metrology 3 , imaging 4 and quantum optics 5 . With its advantages of small footprint, large nonlinearity and dispersion-engineering capability, optical integrated waveguides are excellent candidates for realizing high-gain and large-bandwidth four-wave mixing for which anomalous dispersion is a key condition. Various waveguides based on, for example, silicon, aluminium gallium arsenide and nonlinear glass have been studied 6–10 , but suffer from considerable gain and bandwidth reductions, as conventional design approaches for anomalous dispersion result in multi-mode operation. We present a methodology for fabricating nonlinear waveguides with simultaneous single-mode operation and anomalous dispersion for ultra-broadband operation and high-efficiency four-wave mixing. Although we implemented this in silicon nitride waveguides, the design approach can be used with other platforms as well. By using higher-order dispersion, we achieved unprecedented amplification bandwidths of more than 300 nm in these ultra-low-loss integrated waveguides. Penalty-free all-optical wavelength conversion of 100 Gbit s −1 data in a single optical channel of over 200 nm was realized. These single-mode dispersion-engineered nonlinear waveguides could become practical building blocks in various nonlinear photonics applications.

Four-wave-mixing (FWM) based wavelength conversion offers a versatile solution for flexible optical networks. Significant research has focused on developing new materials and structures to enhance FWM performance. However, high-efficiency waveguide-based wavelength converters typically require either long interaction lengths or high pump power, which in turn limits their phase-matching bandwidth and complicates integration with on-chip lasers. Although microring resonator-based wavelength converters can substantially improve conversion efficiency (CE), they suffer from reduced signal bandwidth due to the resonance filtering effect. In this work, we propose a Fabry–Perot Bragg grating cavity-based singly resonant FWM scheme to enhance CE without compromising the signal bandwidth. In this configuration, only the pump light is resonantly enhanced within the cavity, while the signal and idler light undergo a single pass. We achieve CE enhancements of 16.7 dB in such a cavity compared to a waveguide with the same length on the AlGaAs-on-insulator platform with a phase-matching bandwidth exceeding 200 nm. We also demonstrate a continuously tunable wavelength conversion system, showcasing its potential to support high signal data rates. Our approach provides a promising pathway for on-chip laser-driven nonlinear signal processing, enabling efficient high-speed wavelength conversion and applications where both high CE and data rate are crucial.

We present a flexible generation of various high-frequency analog signals with excellent wavelength scalability using a nonlinear Mach-Zehnder modulator enabled by a compact silicon nitride chip with 4π nonlinear phase shifts for the first time.

Vertical cavity surface-emitting laser (VCSEL)-based optical interconnects (OI) are crucial for high-speed data transmission in data centers, supercomputers, and vehicles, yet their performance is challenged by harsh and fluctuating thermal conditions. This paper addresses these challenges by integrating an ordinary differential equation (ODE) solver within the VCSEL communication chain, leveraging the adjoint method to enable effective gradient-based optimization of pre-equalizer weights. We propose a machine learning (ML) approach to optimize feed-forward equalizer (FFE) weights for VCSEL transceivers, which significantly enhances signal integrity by managing inter-symbol interference (ISI) and reducing the symbol error rate (SER).

Narrow-linewidth yet tunable laser oscillators are one of the most important tools for precision metrology, optical atomic clocks, sensing, and quantum computing.Commonly used tunable coherent oscillators are based on stimulated emission or stimulated Brillouin scattering; as a result, the operating wavelength band is limited by the gain media.Based on nonlinear optical gain, optical parametric oscillators (OPOs) enable coherent signal generation within the whole transparency window of the medium used.However, the demonstration of OPO-based Hertz-level linewidth and tunable oscillators has remained elusive.Here, we present a tunable coherent oscillator based on a multimode coherent OPO in a high-Q microresonator, i.e., a microcomb.Single-mode coherent oscillation is realized through self-injection locking (SIL) of one selected comb line.We achieve coarse tuning up to 20 nm and an intrinsic linewidth down to sub-Hertz level, which is three orders of magnitude lower than the pump.Furthermore, we demonstrate that this scheme results in the repetition rate stabilization of the microcomb.These results open exciting possibilities for generating tunable coherent radiation where stimulated emission materials are difficult to obtain, and the stabilization of microcomb sources beyond the limits imposed by the thermorefractive noise in the cavity.

Frequency locking of lasers is fundamental to a vast number of applications within the field of optics. Usually, when locking a laser to an optical reference wave, it is imperative that locking can be maintained in spite of low reference powers. Previous solutions to frequency locking involve injection locking and/or optical phase locked loops. While previous works have shown locking to weak waves, we extend the lowest demonstrated optical power locked to by approximately 20 dB, realizing locking down to -90 dBm, using a novel digital dither optical phase-locked loop. Measurements of the locked laser phase error verify the performance. The loop design circumvents the presence of a dither on the locked laser light, hence avoiding dither penalties, and low-power locking is realized via coherent detection gain without any optical amplifiers. Low phase noise standard deviations of less than 20°at -80 dBm optical power and Allan deviation of $3\cdot 10^{-16}$ at 1 s averaging time indicate great potential for a variety of applications within optical sensing, communications, and metrology.

Optical networks have long played a central role in telecommunication networks, forming the fiber backbone of the internet. Over time, fiber optic systems have evolved and found deployment increasingly closer to the network edge. Today, optical systems extend to the server network interface cards and home access networks. New application areas have emerged, such as the use of free space communications using LiFi technologies, space communication networks between satellites, and ground stations. Looking ahead, optical systems in many areas will continue to be driven by the need for higher speeds and capacity to keep up with traffic demands. In addition to faster interface speeds, parallel fibers or spatial division multiplexing will be used for future capacity growth. In several application areas, new functionality is expected, such as low latency in XHaul networks and optical switching and co-packaged optics in data centers. LiFi will become critical for mitigating RF interference for in-building networks. Intense research is underway to develop quantum networks to connect quantum computers. This general trend toward new functionalities for optical systems, moving beyond capacity growth in fiber networks, is driven in large part by the increasing performance and demands of today's user equipment and applications. From the network edge to the data centers, components are reliant on optics. However, many of these developments are occurring quite independently and this situation carries the risk of creating problems down the road when eventually all of these components need to be seamlessly connected to maximaze efficiency. Therefore, integration of optics into these new applications and the higher levels of functionality demanded of optics motivate the use of roadmaps to guide research and development to overcome future roadblocks.

The use of long pseudo-random bit sequences to emulate random data in optical transmission links often gives rise to degradation of the receiver sensitivity for a given target bit error rate resulting in a power budget penalty. This is a well know fact and conventionally attributed to the non-ideal response of the system at very low frequencies. Here, we investigate this in the context of directly-modulated vertical-cavity surface emitting laser (VCSEL) based optical interconnects. We observe, both in simulations and in experiments, a substantial increase in the penalty with long sequences especially at symbol rates exceeding the VCSEL bandwidth. Our results show that this is related to the high frequency dynamic behavior of the VCSEL itself and we identify a few certain short bit sequences that cause the large penalty observed. We also show that the penalty can be significantly reduced by a simple modification of the sequences.

Narrow-linewidth yet tunable laser oscillators are one of the most important tools for precision metrology, optical atomic clocks, sensing and quantum computing. Commonly used tunable coherent oscillators are based on stimulated emission or stimulated Brillouin scattering; as a result, the operating wavelength band is limited by the gain media. Based on nonlinear optical gain, optical parametric oscillators (OPOs) enable coherent signal generation within the whole transparency window of the medium used. However, the demonstration of OPO-based Hertz-level linewidth and tunable oscillators has remained elusive. Here, we present a tunable coherent oscillator based on a multimode coherent OPO in a high-Q microresonator, i.e., a microcomb. Single-mode coherent oscillation is realized through self-injection locking (SIL) of one selected comb line. We achieve coarse tuning up to 20 nm, and an intrinsic linewidth down to sub-Hertz level, which is three orders of magnitude lower than the pump. Furthermore, we demonstrate that this scheme results into repetition rate stabilization of the microcomb. These results open exciting possibilities for generating tunable coherent radiation where stimulated emission materials are difficult to obtain, and the stabilization of microcomb sources beyond the limits imposed by the thermorefractive noise in the cavity.

Optical injection locking generally occurs when light from a master laser is unidirectionally injected into a slave laser, such that the injected light overcomes spontaneous emission inside the cavity, and forces the slave laser to behave as a frequency copy of the master. Here, we study the limits of stability for optically pre-amplified optical injection locking in the case of large added noise on the input field and in the presence of a phase locked loop which minimizes the frequency offset between master and slave lasers. We present a set of modified rate equations which we use to describe the physics of the system near the limit of stable injection locking, and report on phase slips which occur due to injected noise momentarily destabilizing the system. We then provide experimental evidence to support the behavior seen in simulation, and are able to successfully recover a CW wave at -80 dBm black box input power (-70 dBm for phase slip free operation), providing 20 dBm of output power from the injection locked slave laser.

Conventional optical amplifiers that use stimulated emission suffer from the generation of excess noise, thus limiting the performance in many applications. The phase-sensitive optical parametric amplifier, relying on the use of a nonlinear material for amplification, is an exception that can approach a noise figure of 0 dB. Its implementation in optical communication links has, however, been cumbersome due to increased complexity both in the transmitter and the receiver, effectively limiting the use of such amplifiers in practice. Here, we propose and demonstrate an implementation of a transmission system with exceptional performance in terms of receiver sensitivity (0.9 photons per bit) using a standalone ultralow-noise phase-sensitively preamplified receiver and a conventional single-wave optical transmitter. This is a significant simplification compared to previous demonstrations and can transform such amplifiers from a curiosity to practical use for example in deep-space-to-earth communication links.

High data rate optical deep space communication links for future space missions need large capture area receivers that can efficiently couple light into a single-mode fiber. Coherent detection is attractive as it offers both high spectral efficiency and sensitivity. Here, we numerically investigate two such large area receivers in the context of weak signal reception; the multi-aperture array and the multi-mode fiber-coupled receiver, together with optical coherent combining. We find that the number of speckles captured by the aperture should match the number of modes supported by the receiver-fiber for high efficiency and sensitivity. Using an optically preamplified dither-optical phase locked loop for tip-tilt, phase, and amplitude compensation, we predict that efficient reception of signals can be maintained down to -80 dBm of received power per mode for realistic atmospheric channels.

Four-wave mixing has extensively been investigated for various applications such as communications, spectroscopy, metrology, quantum computing and bio-imaging. However, there is a clear desire to implement these functionalities in a small footprint nonlinear platform, being capable of efficient operation across a large optical bandwidth. Many such integrated platforms have been explored, but suffer from intrinsic significant performance degradation, because conventional approaches of nonlinear photonic waveguide geometry construction for dispersion engineering focus on waveguide cross section and result in always being multimode as a byproduct. Here we propose and demonstrate a methodology that utilizes not only the impact of the waveguide cross section on the modal and dispersion behavior of the waveguide but also includes the impact of the waveguide bend for cutting off high-order modes. This approach results in simultaneous single-mode operation and dispersion engineering for very broadband operation of four-wave mixing. While we implemented this in silicon nitride waveguides, which has emerged as a promising platform capable of continuous-wave optical parametric amplification, the design approach can be universally used with other platforms as well. By also considering both second- and fourth-order dispersion we achieve unprecedented amplification bandwidths of approximately 300 nm in super-low-loss silicon nitride nonlinear waveguides. In addition, penalty-free all-optical wavelength conversion of 100 Gbit/s data in a single optical carrier over 200 nm is realized, for the first time, without optical amplification of signal or idler waves. These single-mode hyper-dispersion-engineered nonlinear integrated waveguides can become practical building blocks in versatile nonlinear photonic devices and optical networks.