Optical network nodes
A point-to-point fiber link connects two endpoints, while a transport network contains many interconnected links. Nodes terminate, pass, add, drop, or switch optical channels at those interconnection points. These devices are called optical network nodes.
Depending on the node type, a node can regenerate a degraded signal, select wavelengths for local add and drop, or switch optical connections between ports. The functions are implemented by different devices and modules, so a node does not necessarily perform every operation. Together, these functions connect fiber links into a managed communication network.
Signal regenerators and optical reach
As an optical signal travels through a long fiber span, its power decreases through . Its waveform can also be altered by and noise. A regenerator is a node that restores a signal when the accumulated impairments exceed the line system or receiver margin.
Opto-electronic (O-E-O) regeneration
A widely deployed form of regeneration uses an opto-electronic conversion with three functional stages:
- Optical-to-electrical (O/E) conversion: The incoming optical signal is detected and converted into an electrical signal, for example with a photodiode and receiver circuitry.
- 3R regeneration in the electrical domain:The electrical signal is processed using the three functions known as "3R":
- Re-amplifying: The signal amplitude is restored to the level required by the electrical processing stage.
- Re-shaping: A decision and signal-processing circuit reconstructs the intended symbol waveform from the received signal.
- Re-timing: Clock recovery estimates symbol timing and reduces timing errors before the new signal is generated. It does not eliminate every source of jitter.
- Electrical-to-optical (E/O) conversion: The processed electrical signal drives an optical source or modulator, producing a new optical signal for the next fiber span.
O-E-O regeneration restores amplitude, waveform, and timing, and it requires electronic processing for the signal format. The supported rate, modulation, and number of channels are limited by the installed electronics and can constrain a high-capacity node.
All-optical signal processing
All-optical signal processing has been studied as a way to perform selected regeneration functions without an O-E-O conversion. Demonstrations use devices with a nonlinear optical response, including a nonlinear optical gate. The available regeneration functions depend on the signal format and node design, so complete 3R performance must be verified for the specific equipment.
A nonlinear gate can have an "S-shaped" power-transfer characteristic over a selected operating range. Depending on the device and bias point, it can reduce low-level fluctuations and limit the output of high-level symbols. This may improve waveform quality in the optical domain, but the transfer curve, noise performance, and usable rate are device dependent.
WDM nodes: OADM and OXC
In networks using , several optical carriers share one fiber. Network nodes add or drop selected carriers and can switch optical connections between input and output ports.
OADM (Optical Add-Drop Multiplexer): local add and drop
An OADM selectively removes (drops) one or more wavelengths from a WDM signal and inserts (adds) local data on those or other available wavelengths. The remaining channels pass through the device subject to insertion loss, filtering, and crosstalk specifications.
This creates an access point on a long-haul link or metropolitan ring, where local traffic can be added or dropped while other channels continue along the line. A reconfigurable OADM, commonly called a ROADM, uses remotely controlled wavelength-selective elements for this function. Wavelength selection controls the add and drop operation, while end-to-end routing also depends on the network topology, node configuration, and control plane.
OXC (Optical Cross-Connect): optical switching node
An OXC is an optical switching matrix that connects optical input ports to output ports. Depending on its architecture, it can switch individual wavelengths, groups of wavelengths, or complete fiber streams. The switching granularity and supported flex-grid functions are properties of the equipment.
OXCs are used in core and transport nodes. An OXC can:
- Connect major network segments (e.g., interconnecting multiple WDM rings).
- Dynamically reconfigure network topology to adapt to changing traffic patterns.
- Provide protection and restoration by rerouting traffic around failures, with the response time determined by the switching and control architecture.
Requirements for optical network nodes
For OADMs and OXCs to function reliably, the node design must meet optical, operational, and scaling requirements:
- Low insertion loss and crosstalk: The node should stay within the signal-power budget. Crosstalk, meaning leakage from one channel into another, must remain within the limit specified for the modulation and receiver.
- Polarization performance: The node and its coherent interfaces must meet the required polarization-dependent performance for the line system. The polarization state can vary during propagation, so the design must account for that variation.
- Reliability and scalability: As infrastructure elements, nodes need a defined availability target and maintainable components. The design should allow ports, channels, or degrees to be added within the platform limits.
- Switching time: OXC switching speed must match the intended provisioning, restoration, or protection function. The response can range from control-plane times to faster hardware protection times, depending on the technology.
- Remote management: Transport deployments commonly provide management functions for configuring, monitoring, and diagnosing nodes through an NMS. The available operations and degree of automation depend on the vendor platform.