Terabit network becomes possible
Researchers from Australia, Denmark and China have jointly proved the possibility of a terabit (1000 gigabit or 1,000,000 megabit) data network via optical cables. The solution uses a photon processor made of exotic material, chalcogenide, for signal processing.
The results of the group's work were published on February 16, 2009 in an article in the journal Optics Express. It describes the demonstration of 640 GB network and the expansion of this approach to achieve speeds of 1TB.
')
Ben Eggleton, director of the Australian Center for Ultrasonic Devices for Optical Systems (Center for Ultrahigh bandwidth Devices for Optical Systems [CUDOS]) said that the problem is not entering such a large amount of high-speed data into the optical core, but in receiving data from there at such speeds.
According to Eggleton, lasers driven by familiar electronics can inject dozens of 10 Gbit streams, but electronics are not fast enough to get multiplexed data at speeds above 40 Gbit / s.
Together with a Danish research organization working on high-speed optical networks, CUDOS is developing a photon integrated circuit that uses lasers and light just like ordinary electronics uses electrons and transistors.
To seal the channel, time division signals are used. The modulation of separate, closely spaced wavelengths makes it possible to obtain a rather large increase in comparison with current systems that can transmit gigabits on each band of the spectrum that is far from the other.
One of the key discoveries made by researchers is not even about speed, but about practicality. The use of traditional methods for obtaining microcircuits from chalcogenide, arsenic trisulfide (As2S3) made it possible to reduce the length of the demultiplexer channel from tens of meters to only five centimeters.
According to Eggleton, the non-linearity of the material is the key to reducing the length of the waveguide, and this has become possible thanks to the research of materials produced by the CUDOS group. He said: “We found a material whose non-linearity is three orders of magnitude higher than that of previous attempts. This allows us to achieve the same results on a miniature crystal, instead of optical cable coils, that were used in previous methods of achieving high transmission speeds. ”
Designing a monolithic crystal that can process 640 Gbps and more requires a separate waveguide for each of the 10 Gbps transmission channels.
It is reported that silicon chips can also use this principle to achieve somewhat smaller results, but the main goal of the group was to create a fully photonic microcircuit, in the same factories that CMOS are currently producing.
It may take years to get practical results from these studies, but these demonstrations are already asking for serious suggestions.
The results of the group's work were published on February 16, 2009 in an article in the journal Optics Express. It describes the demonstration of 640 GB network and the expansion of this approach to achieve speeds of 1TB.
')
Ben Eggleton, director of the Australian Center for Ultrasonic Devices for Optical Systems (Center for Ultrahigh bandwidth Devices for Optical Systems [CUDOS]) said that the problem is not entering such a large amount of high-speed data into the optical core, but in receiving data from there at such speeds.
According to Eggleton, lasers driven by familiar electronics can inject dozens of 10 Gbit streams, but electronics are not fast enough to get multiplexed data at speeds above 40 Gbit / s.
Together with a Danish research organization working on high-speed optical networks, CUDOS is developing a photon integrated circuit that uses lasers and light just like ordinary electronics uses electrons and transistors.
To seal the channel, time division signals are used. The modulation of separate, closely spaced wavelengths makes it possible to obtain a rather large increase in comparison with current systems that can transmit gigabits on each band of the spectrum that is far from the other.
One of the key discoveries made by researchers is not even about speed, but about practicality. The use of traditional methods for obtaining microcircuits from chalcogenide, arsenic trisulfide (As2S3) made it possible to reduce the length of the demultiplexer channel from tens of meters to only five centimeters.
According to Eggleton, the non-linearity of the material is the key to reducing the length of the waveguide, and this has become possible thanks to the research of materials produced by the CUDOS group. He said: “We found a material whose non-linearity is three orders of magnitude higher than that of previous attempts. This allows us to achieve the same results on a miniature crystal, instead of optical cable coils, that were used in previous methods of achieving high transmission speeds. ”
Designing a monolithic crystal that can process 640 Gbps and more requires a separate waveguide for each of the 10 Gbps transmission channels.
It is reported that silicon chips can also use this principle to achieve somewhat smaller results, but the main goal of the group was to create a fully photonic microcircuit, in the same factories that CMOS are currently producing.
It may take years to get practical results from these studies, but these demonstrations are already asking for serious suggestions.
Source: https://habr.com/ru/post/52037/
All Articles