The research †˜Next generation optical high speed networks' considers new paradigms for optical access and long-haul (1000 km and above) communications, to satisfy the increasing demand for capacity, spectral efficiency and quality of service in telecom networks. In long-haul transmissions, digital signal processing (DSP), with high-order modulation and coherent detection, has revolutionized optical systems, thanks to its effectiveness to digitally mitigate chromatic dispersion and polarization mode dispersion, with an increase in transmission length. However, the drastic increase in system complexity creates a severe challenges for real-time implementation, related also to limited bandwidth of digital to analog converters (DAC), high power consumption of electronic devices, massive use of hardware parallelization and pipeline, making the system intolerant to any feedback path. On the other hand, future access optical passive networks (PON) need to increase per-user bandwidth, evolving towards wavelength-division multiplexing (WDM) systems, with colourless (wavelength independent) low-cost transmitters at the user's side, also allowing different operators to share the same infrastructure. For both core and access networks, orthogonal frequency division multiplexing (OFDM) is proposed to overcome the electronic bottleneck, reduce power consumption and increase the spectral efficiency, satisfying current large bandwidth requests. During the three-year research work, within the EU ASTRON and the Italian ROAD-NGN projects, new optical network architectures, systems and subsystems have been proposed and their performances have been analytically, numerically and experimentally evaluated. In particular, new digital OFDM systems for unbundling the local loop (ULL) in access networks have been investigated for ultra-high speed PONs, along with discrete multitoned (DMT) approaches, based on reflective semiconductor optical amplifier (RSOA). All optical OFDM (AO-OFDM) systems have been experimentally tested during a 6-month staying at the National Institute of Information and Communications Technology (NICT) in Japan, using both coherent and direct detection receivers. The performance of two different optical devices, able to demultiplex the AO-OFDM signal, have been experimentally compared, without optical dispersion compensation or time gating. For long haul transmissions, AO-OFDM can also reduce the power consumption in case of superchannels, with a high number of sub-channels. Analytically formula have been found for performance evaluation, in terms of maximum number of spans, frequency offset, phase noise and equalization enhancement phase noise (EEPN), supported and validated by numerical simulation. Finally, a novel transmission system is proposed, based on the digital fractional Fourier transform, to generate sinc-shaped time domain pulse at the receiver side, reducing the DSP complexity for real time implementation at high data rates.
Next generation optical high-speed network
2016
Abstract
The research †˜Next generation optical high speed networks' considers new paradigms for optical access and long-haul (1000 km and above) communications, to satisfy the increasing demand for capacity, spectral efficiency and quality of service in telecom networks. In long-haul transmissions, digital signal processing (DSP), with high-order modulation and coherent detection, has revolutionized optical systems, thanks to its effectiveness to digitally mitigate chromatic dispersion and polarization mode dispersion, with an increase in transmission length. However, the drastic increase in system complexity creates a severe challenges for real-time implementation, related also to limited bandwidth of digital to analog converters (DAC), high power consumption of electronic devices, massive use of hardware parallelization and pipeline, making the system intolerant to any feedback path. On the other hand, future access optical passive networks (PON) need to increase per-user bandwidth, evolving towards wavelength-division multiplexing (WDM) systems, with colourless (wavelength independent) low-cost transmitters at the user's side, also allowing different operators to share the same infrastructure. For both core and access networks, orthogonal frequency division multiplexing (OFDM) is proposed to overcome the electronic bottleneck, reduce power consumption and increase the spectral efficiency, satisfying current large bandwidth requests. During the three-year research work, within the EU ASTRON and the Italian ROAD-NGN projects, new optical network architectures, systems and subsystems have been proposed and their performances have been analytically, numerically and experimentally evaluated. In particular, new digital OFDM systems for unbundling the local loop (ULL) in access networks have been investigated for ultra-high speed PONs, along with discrete multitoned (DMT) approaches, based on reflective semiconductor optical amplifier (RSOA). All optical OFDM (AO-OFDM) systems have been experimentally tested during a 6-month staying at the National Institute of Information and Communications Technology (NICT) in Japan, using both coherent and direct detection receivers. The performance of two different optical devices, able to demultiplex the AO-OFDM signal, have been experimentally compared, without optical dispersion compensation or time gating. For long haul transmissions, AO-OFDM can also reduce the power consumption in case of superchannels, with a high number of sub-channels. Analytically formula have been found for performance evaluation, in terms of maximum number of spans, frequency offset, phase noise and equalization enhancement phase noise (EEPN), supported and validated by numerical simulation. Finally, a novel transmission system is proposed, based on the digital fractional Fourier transform, to generate sinc-shaped time domain pulse at the receiver side, reducing the DSP complexity for real time implementation at high data rates.I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/273421
URN:NBN:IT:UNIROMA3-273421