We successfully transmitted a net 582-Gb/s probabilistically shaped PAM12 C-band signal over 11-km dispersion-shifted fibre and net 4×526-Gb/s uniform PAM8 O-band signals over 2-km four-core fibre using a single-carrier 216-GBd IMDD system based on a 150-GHz bandwidth InP-DHBT electrical mixer and a thin-film lithium-niobate modulator.

We demonstrate a net bitrate of 633-Gb/s back-to-back and 628-Gb/s 11-km IMDD transmission with 224-GBd PS-PAM14 signal using upper-sideband gain-enhanced mode of a 150-GHz-bandwidth electrical mixer for frequency-domain multiplexing, achieving the first >600-Gb/s/lane IMDD transmission.

We demonstrate 200-Gbaud PAM4 transmission with 10-dB bandwidth of 68 GHz in which an advanced MLSE with simple-soft-output scheme improves NGMI. We show that the simple-soft-output scheme is applicable to turbo product codes.

We developed a 100-GHz-class bandwidth TEC-less O-band 8-channel InP MZ modulator. The small device exhibits net 3.2-Tb/s (400-Gb/s/lane) 500 m transmission over the range of 20 to 80°C.

We demonstrated 232-264-GBd IMDD signal generation and detection using electrical bandwidth multiplexing and demultiplexing configuration based on in-house 150-GHz InP-DHBT mixers with an adaptive reconstruction technique, achieving a record net bitrate per wavelength of 660-Gb/s back-to-back and 651-Gb/s 11-km transmission with 248-GBd PS-PAM12 signals. (c) 2025 The Author(s)

We demonstrate, for the first time, a capacity of 1.6 Tb/s over 2 km of single-mode fiber on the O-band LAN-WDM grid with 4-lane 400-Gb/s/lane signals with 155-GBd PAM-8 signals enhanced by NL-MLSE.

We propose an analog filterless InP-DHBT AMUX-based bandwidth tripler with a time-interleaved nonlinear digital pre-distortion for tripler and optical frontend impairments, achieving a net-bitrate 496.9-Gb/s signal generation and 483.9-Gb/s 11-km transmission with single-carrier 216-GBd PS-PAM8.

We propose a simple LLR-calculation method which modifies the LLR distribution using hard-decision information for IM-DD systems with MLSE and SD-FEC. The proposed method achieves high NGMI in 128-Gbaud PAM4 transmission with 40-GHz bandwidth limitation.

To economically manage the rapid traffic growth in data center networks, transmission technologies need to be studied for next-generation high-speed Ethernet, such as 1.6 TbE and beyond. This article describes a demonstration of 1.6 Tb/s (4 x 400-Gb/s/lane) O-band transmission over 10 km of installed four-core fiber. Since chromatic dispersion limits the transmission distance of the high-speed intensity-modulated direct detection (IM-DD) signals, a space division multiplexed technology using the 10-km four-core fiber is suitable to parallelly transmit 400-Gb/s/lane signals at near-zero dispersion wavelengths. A net-rate of 400 Gb/s/lane IM-DD signals with 155-GBd pulse amplitude modulation-8 (PAM-8) is generated by using an in-house broadband amplifier based on an InP double hetero-junction bipolar transistor (InP-DHBT). Our nonlinear maximum likelihood sequence estimation (NL-MLSE) enhances the performance of the 400-Gb/s/lane signals. We also introduce a technique called trellis path-limitation MLSE (TL-MLSE) for reducing computational complexity with temporarily decided results and a truncated trellis diagram. The trellis path-limitation MLSE with nonlinear calculation function (TL-NL-MLSE) achieves 1.6-Tb/s 10-km transmission in the O-band with the same performance as and lower computational complexity than the NL-MLSE.

The chromatic dispersion tolerance of 400-Gb/s/lane net-rate intensity-modulated direct-detection signals in the O-band is verified through 20-km single-mode fiber transmission, showing that 1.6-Tb/s 2-km transmission with 4 lanes of wavelength division multiplexing and 1.6-Tb/s 10-km transmission with 4 lanes of space division multiplexing are feasible.

This paper provides an overview of the trends in high-capacity IM-DD transmission in the O-band. We also present our recent research results of 400-Gb/s/lane transmission over installed 10-km 4-core fiber, achieving a 1.6-Tb/s total capacity. (c) 2024 The Author(s)

Colorless, directionless, and contentionless reconfigurable optical add/drop multiplexing (CDC-ROADM) provides highly flexible physical layer network configuration. Such CDC-ROADM must operate in multiple wavelength bands which are being increasingly implemented in optical transmission systems. The operation in C+L bands requires switch devices used in CDC-ROADM to also be capable of multiband operation. Recent studies on wavelength division multiplexing (WDM) systems have pointed out the impact of amplified spontaneous emission (ASE) noise generated by signals of different wavelengths, which causes OSNR degradation. Therefore, it is desirable to filter out the ASE noise from different transponders when multiplexing multiple wavelengths at the transmitter side, especially in a system with non-wavelength selective combiners such as directional couplers and multicast switches. The use of transponder aggregators with filtering functions, such as the M x N wavelength selective switch (WSS), is preferable for this filtering. However, the downside of these devices is that it is difficult to provide economical multiband support. Therefore, we propose an economical transponder aggregator configuration by allowing a certain amount of ASE superposition and reducing the number of filtering functions. In this paper, we fabricated a prototype of the proposed transponder aggregator by combining silica-based planar lightwave circuit technology and C+L band WSS, both commercially available, and verified its feasibility through transmission experiments. The novel transponder aggregator is a practical solution for a multiband CDC-ROADM system with improved OSNR performance.

We propose a simple method to calculate LLR for IM-DD system with MLSE and SD-FEC. We show that an advanced MLSE with the simple calculation method which shapes the LLR distribution makes NGMI higher for 128-Gbaud PAM4 signal in a severe bandwidth limitation.

We propose a transceiver configuration with optical pre- and post-equalization for channel distortion and show that the configuration enhances the performance of 500- Gbps/lambda PDM-32QAM real-time transmission and frequency-offset tolerance under a severe bandwidth limitation.

We propose advanced maximum likelihood sequence estimation (MLSE) methods for increasing the capacity of short-reach intensity-modulated direct detection (IM-DD) transmissions for economically realizing next generation Ethernet such as 800 GbE or 1.6 TbE. It is particularly important to conquer the traffic growth in data center networks, which has become a major issue. In this article, we introduce a method for reducing the complexity of the Viterbi algorithm, which is a problem in the actual implementation of MLSE, by using a penalty-less truncated trellis diagram with temporal decision results. Next, we introduce a method for improving the demodulation performance by precisely emulating the nonlinear response of the transmission system with low-cost devices using a nonlinear filter. Finally, we introduce a method for reducing the computational complexity with a decision feedback function. Transmission experiments in the O-band with transmission rates of 200 Gbps/lane or higher using each of the proposed methods are also presented. Finally, we present a 200 Gbps/lane 4-LAN-WDM 800 G 2 km transmission experiment using a 4-channel LAN-WDM TOSA designed for 400 G transmission.

We proposed and experimentally demonstrated a trellis-path-limited sequence estimation method combining the log-likelihood ratio from minimum-mean-square-error equalizer-enabled signal detection with soft-decision -FEC decoding under severe bandwidth limitations, achieving 1.35-Tb/s 256-GBd 16QAM and 1.65-Tb/s 208-GBd 64QAM.

Record 400-Gb/s/lane net-rate (155-GBd PAM-8) IM-DD signals are transmitted in the O-band over installed 10-km 4-core fibre using in-house broadband InP-DHBT amplifier and our nonlinear-MLSE, achieving a 1.6-Tb/s total capacity for future data-centre Ethernet networks.

We report four-wave mixing mitigation by 4λ-WDM filters fabricated by silica-based planar lightwave circuits for 800- and 400-GbE application. The multiplexing filter integrates a simple polarization-alternating structure on arbitrary lanes for the impairment mitigation.