As we know, since the 1990s, WDM WDM technology has been used for long-haul fibre-optic links of hundreds or even thousands of kilometres. For most regions of the country, the fibre infrastructure is its most expensive asset, while the cost of transceiver components is relatively low.
However, with the explosion of data rates in networks such as 5G, WDM technology is becoming increasingly important in short-haul links as well, which are deployed in much larger volumes and are therefore more sensitive to the cost and size of transceiver assemblies.
Currently, these networks still rely on thousands of single-mode optical fibres transmitted in parallel through channels of space division multiplexing, with relatively low data rates of at most a few hundred Gbit/s (800G) per channel, with a small number of possible applications in the T-class.
However, in the foreseeable future, the concept of common spatial parallelisation will soon reach the limits of its scalability, and will have to be complemented by spectral parallelisation of the data streams in each fibre in order to sustain further increases in data rates. This may open up a whole new application space for WDM technology, in which maximum scalability in terms of number of channels and data rate is crucial.
In this context, the optical frequency comb generator (FCG) plays a key role as a compact, fixed, multi-wavelength light source that can provide a large number of well-defined optical carriers. In addition, a particularly important advantage of optical frequency combs is that the comb lines are intrinsically equidistant in frequency, thus relaxing the requirement for inter-channel guard bands and avoiding the frequency control that would be required for a single line in a conventional scheme using an array of DFB lasers.
It is important to note that these advantages apply not only to WDM transmitters but also to their receivers, where discrete local oscillator (LO) arrays can be replaced by a single comb generator. The use of LO comb generators further facilitates digital signal processing for WDM channels, thereby reducing receiver complexity and increasing phase noise tolerance.
In addition, the use of LO comb signals with phase-locking for parallel coherent reception even makes it possible to reconstruct the time-domain waveform of the entire WDM signal, thus compensating for impairments caused by optical nonlinearities in the transmission fibre. In addition to these conceptual advantages of comb-based signal transmission, smaller size and cost-effective mass production are also key for future WDM transceivers.
Therefore, among the various comb signal generator concepts, chip-scale devices are of particular interest. When combined with highly scalable photonic integrated circuits for data signal modulation, multiplexing, routing and reception, such devices may hold the key to compact, highly efficient WDM transceivers that can be fabricated in large quantities at low cost, with transmission capacities of up to tens of Tbit/s per fibre.
The following figure depicts a schematic of a WDM transmitter using an optical frequency comb FCG as a multi-wavelength light source.The FCG comb signal is first separated in a demultiplexer (DEMUX) and then enters an EOM electro-optical modulator. Through, the signal is subjected to advanced QAM quadrature amplitude modulation for optimal spectral efficiency (SE).
At the transmitter egress, the channels are recombined in a multiplexer (MUX) and the WDM signals are transmitted over single mode fibre. At the receiving end, the wavelength division multiplexing receiver (WDM Rx), uses the LO local oscillator of the 2nd FCG for multiwavelength coherent detection. The channels of the input WDM signals are separated by a demultiplexer and fed to the coherent receiver array (Coh. Rx). where the demultiplexing frequency of the local oscillator LO is used as a phase reference for each coherent receiver. The performance of such WDM links obviously depends to a large extent on the underlying comb signal generator, in particular the optical line width and the optical power per comb line.
Of course, optical frequency comb technology is still in the developmental stage, and its application scenarios and market size are relatively small. If it can overcome technical bottlenecks, reduce costs and improve reliability, then it will be possible to achieve scale-level applications in optical transmission.
Post time: Nov-21-2024