400G DWDM CFP2-DCO Module

The growth of network traffic leads to increased port bandwidth on the transmission network. For long-distance and high-bandwidth transmission, wavelength-division multiplexing (WDM)-based coherent transmission technology provides the best solution.

As 400G coherent solutions mature, the demand for 400G coherent ports will proliferate. There are two drivers for the growth of 400G coherent ports:

•  Network bandwidth growth;
•  Increase in the number of 400GE ports on the client side.

It is proved to be the most cost-effective method to use one 400G wavelength to carry 400GE traffic.

According to LightCounting’s forecast report, 400G coherent ports will be used in more and more networks and will see the fastest growth over the next 5 years. With the continuous increase of network traffic, total wavelengths, and the number of wavelengths in a single network, network operators will also increase the flexibility requirements of network management and scheduling, thus promoting the large-scale deployment of ROADM (reconfigurable optical add/drop multiplexer).

Internet service providers (ISP)can dynamically configure wavelength paths as needed through wavelength selective switching (WSS) technology. Optical paths can realize point-to-point connections, reducing latency and power consumption. Due to these benefits, more and more ISPs are adopting this solution.

For example, in 2017, one of the Chinese ISPs built ROADM networks with 364 wavelengths along the middle and lower reaches of the Yangtze River. Flexible rate modulation and flexible mesh technology make DWDM networks more flexible and resilient, whereas traditional DWDM systems use a fixed 50/100 GHz mesh, center frequency, and channel width. If flexible modulation and grid technology are available, the modulation format and channel width of each port can be customized according to capacity and transmission distance, improving spectral efficiency and transmission capacity. The following is a schematic diagram of a flexible rate and mesh for flexible network configuration.

Figure1: schematic diagram of flexible rate and grid

Changes in network architecture require more flexible line-side optical modules that support Flex Rate and Flex Grid. The current trend in optical networks is towards higher spectral efficiencies, approaching the Shannon limit. Coherent optical modules are developing in three directions:

• Spectral efficiency: Improve the spectral efficiency and single-fiber capacity according to the progress of the oDSP algorithm;
• Baud rate: Increase the single wavelength baud rate, obtain higher single port bandwidth, and reduce cost and power consumption per bit;
• Smaller size and lower power consumption: It adopts integrated optoelectronic components, an advanced manufacturing process, and a dedicated oDSP algorithm.

Due to the Shannon limitation, the 64 Gbaud 400G wavelength cannot achieve the performance required for optical transmission over long distances. Higher baud rates and more complex and powerful oDSP algorithms are required to meet the requirements of intercity (regional) and long-distance backbone networks.

For example, for long-distance links (> 1000 km), the baud rate for 400G wavelengths should be above 90 Gbaud, and both ADC and DAC rates in the oDSP need to be increased. However, as baud rates increase, fiber optic transmission is more expensive and more difficult to compensate for. Therefore, stronger compensation algorithms are needed to compensate for physical lane damage.

Given that ROADMs have been widely used, an end-to-end wavelength link needs to pass through several or even dozens of ROADMs, including wavelength selective switches (WSS). The superposition effect of WSS filtering reduces the effective bandwidth of the link, which puts forward higher requirements for the compensation algorithm in the oDSP. The figure below shows the effect of multi-stage ROADM on optical channel bandwidth.

Figure2: the effect of multi-stage ROADM on optical channel bandwidth

In addition, many ISPs want to flexibly configure modulation format and baud rate according to port rate and transmission distance. For instance, they deploy 400G 16QAM for 400G long-haul transmission and 800G 64QAM for tens of kilometers of metro data center interconnects to improve spectral efficiency and reduce cost per bit. With this flexible modulation technique and a flexible mesh of optical layers, fiber capacity can be maximized, saving optical cable investment.

FiberMall’s long-distance and large-capacity 400G coherent optical transceiver solutions meet the needs of different customers. Each module supports flexible rate modulation (100G/200G/400G) and is packaged in CFP2. It supports 40nm C-band spectral width and 48nm Super C-band at the same time and supports a maximum of 120 wavelengths to meet the large-capacity needs of customers.

Small-size silicon photonics components or high-performance high-bandwidth InP components are used to meet a range of different application scenarios. The principles of 400G coherent optical modules in different packages are the same. The Tx end of the 400G coherent optical module consists of oDSP, data driver, wavelength tunable laser, and PDM-I/Q modulator.

First, the data from the motherboard is mapped and encoded. The Tx-oDSP then performs spectral shaping and compensation for the data link bandwidth. After that, the data driver amplifies the amplitude and inputs the amplified data to the modulator. The modulator then converts the data into an optical signal for output. On the Rx side, the optical signal enters the ICR and interferes with the wavelength of the local oscillator to realize photoelectric conversion. After the high-speed ADC samples the electrical signal, it compensates for the Chromatic Dispersion (CD) and State of Polarization (SOP). The following is the block diagram of the coherent optical module.

Figure3: Block Diagram of Coherent Optical Module

 

Here are the suggestions for 400G CFP2-DCO optical modules used for 400G large-capacity long-distance transmission:

• Compliant with CFP2 protocol (MSA);
• Using the CFP2 package;
• Compliant with 400G CAUI-8 and FlexO interface specifications;
• Supports multiple modulation formats, including QPSK and 16QAM;
• Supports 400G 16QAM 500 km @ 75 GHz and 200G QPSK 2000 km @ 75 GHz.

The 400G CFP2-DCP is a pluggable optical module that provides optimal performance and incorporates several innovative technologies to enhance 400G transmission performance. The following shows the 400G CFP2-DCO block diagram.

Figure4:Block Diagram of 400G CFP2-DCO

• High-performance and low-power consumption oDSP
• To increase the transmission distance, Turbo Product Codes (TPC) FEC technology – high performance, low power consumption – is adopted to approach the Shannon limit. Elastic rates from 200G to 400G are also supported. In addition, pluggable and low-power features are implemented with a low-power IP/DSP architecture.

For 400G CFP2-DCO, multiple modulation formats are supported, including 400G 16QAM, 200G QPSK, and DQPSK. For high-capacity transmission, 16QAM is recommended for single-wavelength 400G@75GHz transmission. For new networks, it is recommended to use QPSK for 200G@75 GHz transmission with a transmission distance of 2000 km. In contrast, DQPSK is applied to existing networks in mixed scenarios to reduce the impact on linearity.

• Super C-band capability
• In the wavelength division multiplexing system, the capacity of a single fiber system is directly affected by the number of transmission wavelengths. The CFP2 module is the first Super C-band optical module, supporting 80 wavelengths of 400G@75G and having a single-fiber optical transmission capacity of 32T. The realization of the Super C-band relies on other capabilities, including underlying lasers, ICTRs, and built-in optical amplifiers (OAs).

Tx and Rx share a single laser to achieve a low-power compact design in a CFP2 package. Additionally, FiberMall’s unique laser design utilizes a compact nanolaser with high output optical power. Below is an ultra-broadband spectrum (120 wavelengths).

Figure5：Ultra-broadband spectrum

• Large output optical power adjustment range
• In long-distance transmission, the output optical power needs to be fine-tuned to obtain better performance. The output optical power of 400G CFP2-DCO can be precisely adjusted in the range of +1dBm to +4dBm to meet the input power requirements of different optical layers.
• Silicon photonics integrated ICTR
• Silicon photonics ICTR technology is used in the 400G CFP2-DCO module to minimize the physical size. Due to its unique optical properties, silicon photonics has greater optical field confinement, resulting in more compact waveguide structures. What’s more, silicon photonics supports polarization processing, which enables modulation and coherent detection of dual-polarized 16QAM signals while minimizing the ICTR chip size.
• Photoelectric multi-chip packaging

The performance of the RF link from the oDSP to the optical modulator is optimized to reduce driver requirements and thus reduce power consumption. In addition, optical chips and electronic chips are packaged together to reduce physical size.

• High-performance compact OA

Silicon photonics ICTR technology is used to achieve compact size but results in large insertion loss. For the requirements of high-performance optical transmission, the output terminal adopts a small OA independently developed by FiberMall to amplify the optical signal. Furthermore, the NF of the OA is optimized for high-quality amplified optical signals.

Here are the suggestions for the 400G MSA optical module for long-distance and ultra-large-capacity transmission:

• High-performance oDSP
• To increase the transmission distance, high-performance FEC technology is used to continuously approach the Shannon limit, and elastic rates of 200-800G are supported. When the number of ROADMs and the number of cascaded filters in the all-optical network architecture increases, the Faster-Than- Nyquist (FTN) algorithm is used to enhance the pass-through capability of the filters, ensuring that the multi-stage filters do not cause losses. The data acquisition and analysis module of the optical fiber link is integrated into the network management system to improve the operation and maintenance capability during the whole life cycle. The transmission performance of the 400G MSA is shown in the figure below.

Figure6: Transmission Performance of the 400G MSA

• High-performance laser
• In a coherent 400G system, a tunable laser provides an optical signal at Tx for modulation. At Rx, another tunable laser provides the optical signal, which is used as a local reference signal for coherent detection. The laser should have the following characteristics:

– High output optical power: ensure the high incident optical power of the module and improve the transmission performance;

– Narrow linewidth: nonlinear phase noise is introduced after the optical signal transmitting through the optical fiber, and the linewidth is directly related to the phase noise. This is especially true for high quadrature amplitude modulation (QAM) transmissions, which further increase the linewidth requirements. A unique InP integrated laser with SOA is used to ensure high output optical power.

In addition, the unique grating design and wavelength control scheme are used to achieve ultra-narrow linewidth and high-stability wavelength locking. What’s more, by optimizing the gain medium and tunable grating of the laser, a tunable laser in the Super-C band is covered. The picture below shows a high-performance laser.

Figure7： A high-performance laser

• High-performance modulator
• Typically, modulators are created by using one of the technologies: lithium niobate (LiNbO 3 ), indium phosphorous (InP), or silicon photonics. Each has its strengths and weaknesses. LiNbO3 is a mature optical component platform that can achieve high bandwidth and low-drive amplitude, but with a large component size. InP supports high bandwidth modulation and can integrate SOA to achieve high output optical power. However, InP is sensitive to temperature, and temperature control requires TEC.
• On the other hand, silicon photonic modulators integrate polarization multiplexing functional units at the chip level while reducing the physical size, which requires larger driving voltages. The 400G MSA uses a semi-insulating substrate and unique Mach-Zehnder modulator for high bandwidth InP I/Q-MZ and SOA integration. In this way, high modulation bandwidth and high output optical power are achieved. The following figure shows the high modulation bandwidth supported by the InP modulator.

Figure8： the high modulation bandwidth supported by the InP modulator

• High-performance optoelectronic or RFIC
• At the Tx of the coherent optical receiver, a driver is required to amplify the electrical signal to drive the optical modulator. On the Rx side, a TIA is required to convert the current signals to a voltage signal and amplify the voltage signal. Therefore, drivers and TIAs need to have higher bandwidth and better linearity.
• It realizes ultra-high bandwidth, ultra-high linearity, ultra-low noise linear driver, and TIA, based on innovative circuit architecture and active equalization design. Coherent Drive Modulators (CDM) and ICR also provide high bandwidth.

Figure9：TIA and driver

• High-performance ICR
• An ICR is used at Rx to receive the optical signal in a coherent optical receiver, This process also involves optical mixers and PDs that are used to convert optical signals into electrical signals. ICR-related technologies include silicon-on-insulator (SOI) technology for ICR integration, planar lightwave circuit (PLC) technology for optical mixers, and an InP PD.

• Optical mixers based on SiN technology can be used to achieve good fiber coupling and polarization processing to obtain the best optical mixing effect. InP PDs with high bandwidth and high sensitivity are mounted on SiN chips through a unique flip-chip package, which forms high-integration, high-performance, and small-size ICRs. The ICR diagram is as follows.

Figure10： Schematic diagram of ICR

• High-performance package
• The 400G MSA uses a high-performance Charger Device Model (CDM) package. The high-bandwidth driver and modulator are packaged in a single assembly, reducing trace lengths for high-speed RF signals, and ensuring high-speed signal integrity and high bandwidth. Some electrical ports use pins to ensure stable connection and bandwidth of incoming signals, thereby improving the performance of CDM components. The following figure is a schematic diagram of high-performance component packaging.

Figure11： schematic diagram of high-performance component packaging

• 200-800G flexible rates, single-wave 800G large-capacity transmission
• The micro-module supports high-order QAM by powerful oDSP and high-bandwidth optics. Meanwhile, constellation shaping 2.0 is used to support 200-800G adjustment. In addition, the built-in OA can guarantee the output optical power under higher order modulation.

Figure12：Flexible modulation formats

The need for higher capacity, lower cost per bit, and lower power consumption is driving higher and higher transmission rates for optical modules. As the mainstream technology of the previous generation, 100G has entered the mature and stable life cycle, and it’s difficult to reduce the unit cost. At present, mainstream 400G optical modules have been used in various network scenarios such as data center networking, metropolitan integrated bearer networks, and large-capacity long-distance transmission networks.