Abstract
Optical communication systems have continuously evolved to support the demand for increasing transmission capacity. The introduction of digital coherent systems represented a revolution since they allowed the exploitation of not only the amplitude but also the phase and polarization diversity of the light. In addition to the use of additional dimensions, digital coherent systems paved the way for novel and efficient compensation of impairments such as phase noise, chromatic dispersion, and polarization mode dispersion. The mitigation of nonlinear distortion, however, remains unresolved since many approaches require prohibitively high computation complexity. In this context, machine learning has emerged as a high-potential solution to achieve the mitigation of nonlinear distortion without increasing computational complexity to prohibitively high levels. The development of machine learning techniques becomes more challenging as we increase the dimensionality of the system, that is, in high inter-symbol interference scenarios, such as high-bit rate and multi-channel systems. (AU)
| Articles published in Agência FAPESP Newsletter about the research grant: |
| More itemsLess items |
| TITULO |
| Articles published in other media outlets ( ): |
| More itemsLess items |
| VEICULO: TITULO (DATA) |
| VEICULO: TITULO (DATA) |