Aoptix was a company that did much than this more years ago. Founded in 2000 and despite 16 years and $150M raised, they could not find much of a market. The Aoptix system could transmit 4x this Google system, had further reach, and achieved 5 nines reliability.
Very cool. I'd like to better understand the difference between this and other line of sight transmission stuff. I assume bandwidth and latency, but not sure the orders of magnitude.
Microwave links (including point-to-point "line-of-sight" links) transmit a cigar-shaped cone of signal which raises the noise floor for all receivers within earshot of a given frequency at a given location. The absolute physical dimensions of the cone of signal are dictated by the specific frequency being transmitted, the size and shape of the antenna, and the distance from the transmitter. This makes interference (including self-interference) one of the bigger challenges to colocating many links in the same geographical area.
One of the promises of millimeter-wave radios (radios which operate at frequencies from 18Ghz-100+Ghz) is that they are less prone to interference because they operate at such high frequencies.
Presumably, one of the bigger reasons to use free space optics is that it results in an almost total elimination of interference. No interference means no coordination is necessary between operators. No coordination between operators means no licensing/regulations are necessary for maximizing use of the available spectrum. Non-regulated or lightly-regulated technologies can have much broader adoption in the marketplace.
Lots of good reasons to pursue something like this.
Comments
Aoptix was a company that did much than this more years ago. Founded in 2000 and despite 16 years and $150M raised, they could not find much of a market. The Aoptix system could transmit 4x this Google system, had further reach, and achieved 5 nines reliability.
https://www.lightreading.com/mobile/backhaul/exclusive-its-l...
https://www.lightreading.com/mobile/backhaul/aoptix-rejoins-...
The main problem with wireless optical backhaul is the high cost. It cannot compete with RF wireless.
700 TiB / 20 days ~= 3.56 Gbps
700 TB / 20 days ~= 3.24 Gbps
20 years ago, fully open source design, 10Mbit https://en.wikipedia.org/wiki/RONJA
Very cool. I'd like to better understand the difference between this and other line of sight transmission stuff. I assume bandwidth and latency, but not sure the orders of magnitude.
Why use this instead of line-of-sight microwave links? Increased bandwidth at the expense of range?
Microwave links (including point-to-point "line-of-sight" links) transmit a cigar-shaped cone of signal which raises the noise floor for all receivers within earshot of a given frequency at a given location. The absolute physical dimensions of the cone of signal are dictated by the specific frequency being transmitted, the size and shape of the antenna, and the distance from the transmitter. This makes interference (including self-interference) one of the bigger challenges to colocating many links in the same geographical area.
One of the promises of millimeter-wave radios (radios which operate at frequencies from 18Ghz-100+Ghz) is that they are less prone to interference because they operate at such high frequencies.
Presumably, one of the bigger reasons to use free space optics is that it results in an almost total elimination of interference. No interference means no coordination is necessary between operators. No coordination between operators means no licensing/regulations are necessary for maximizing use of the available spectrum. Non-regulated or lightly-regulated technologies can have much broader adoption in the marketplace.
Lots of good reasons to pursue something like this.
What happens on a foggy day?