100x Faster Than Wi-Fi: Light-Based Networking Standard Released

In high density living (condo towers, downtown etc), it is common to have interference from neighbor's wifi network on the same floor or adjacent floors.

Having a network that expands using a non-interfering frequency would be a godsend. Especially given the limited number of channels for existing WiFi standard.

You can encode a lot more information in higher frequencies.

But each frequency has different things that are opaque to it, and travel different distances before dropping off.

Wi-Fi is already “really low frequency, really low intensity infrared light”, so I suspect the article is correct when it says we won’t notice it.

I don't know anything about LiFi, but for EM radiation in general, the higher the frequency, the higher the theoretical maximum bandwidth.

https://en.wikipedia.org/wiki/Shannon%E2%80%93Hartley_theore...

Wifi is 2.6 Ghz or 5Ghz, visible red light is 430 terahertz.

Or it sounds like LiFi is just pulsing the lights on an off, in which case the Nyquist Rate that was invented for telegrams is a better analogy

https://en.wikipedia.org/wiki/Nyquist_rate

My understanding is they went with 800 to 1000 nm (infrared), or ~ 375 to 300 THz. I'm not sure as to the total combined bandwidth of all of WiFi 6 or 7, but 75,000 GHz band gives them a lot to play with.

> Even using infrared, would it interfere with things like baby monitors in night mode?

If you're referring to the infrared LEDs that illuminate the baby, their light is not polarized, while light used for communication is, so a polarizing filter in the receiver can filter out such noise.

I think Li-Fi is more comparable to Ethernet instead of Wi-Fi, not because the fundamentals are very different but because of the link budgets available. If you think of an ethernet line as a pipe carrying data, then a collimated laser can be thought of in a similar way - most of the energy that you transmit is going to make it to the destination, and that is not the case with Wi-Fi, even with very high gain antennas. This allows for different modulation schemes and thus higher throughput. Copper ethernet is now capable of 1.6 Tbps [1] and Li-Fi doesn't seem so very fast compared to that; however keep in mind this is comparing only physical layers. Demonstrations of optical laser links of hundreds of Gbps over hundreds of kilometers have taken place [2] using COTS optical ethernet transceivers and special output stages providing precise collimation and pointing.

[1] https://en.wikipedia.org/wiki/Ethernet_physical_layer#1.6_Tb...

[2] https://ntrs.nasa.gov/api/citations/20210026855/downloads/sp...

For your second question, I'm not sure how baby monitors work but the proposed wavelengths are unlicenced and there are little if any rules for how to deal with interference. There are rules for eye safety of laser which limits the maximum energy that can be delivered to the output. Generally as Li-Fi gets more common we will have to learn to deal with interference as it arrives. For example, Lidar systems (older, noncoherent ones) interfere with one another and are even susceptible to interferece from IR motion detectors and such, but these aspects have to be considered during design.

> Newb question: since they are both electromagnetic radiation, how does Li-Fi get up to 100x faster than Wi-Fi?

They specified that they wanted to be 100x faster, and they added parallel channels, until they reached that goal. No, really, this is the real reason.

It is entirely nonsensical to ask for a physical reason, because different channels are just different.

This is mentioned in the article

> Of course, Li-Fi isn’t going to sweep away Wi-Fi and 5G alternatives (nor wired networks). Radio waves still have a distinct advantage with regard to transmission through the atmosphere at great distance, and though opaque objects. Instead, work must concentrate on using horses for courses – with Li-Fi advantages being harvested where possible.

I’m not sure that this provides any benefit from an EMSEC perspective, as you’d basically be going from a position where you’re avoiding radios and even cables and similar things that aren’t TEMPEST shielded for fear of emissions leaking sensitive data, to a position where you’re broadcasting your sensitive data over the air, and a listening device simply needs to look at your lights somehow. I will agree that it’s easier to block light than radio, but I think that’s where the advantages end.

> E.g. if I put my LiFi enabled phone in my pocket mid download, it will completely cease to work.

The light part, sure, but the regular radio wifi part will be fine; it'll be slower, but it won't go away. Ideally there's seamless transition between the networks, LiFi if you have your phone out and there's a sender in the receiver's signal, WiFi in other cases.

>* if I put my LiFi enabled phone in my pocket mid download, it will completely cease to work.*

As per your EMSEC use-case, this is also a privacy benefit, as your pocket becomes a defacto faraday cage guaranteeing that your devices can only transmit information when you want them to.

In your pocket, or just turn around so your body is in between the source and your device? Maybe reflections get the job done, but that has to harm data rates, right? I am not an optical expert.

> wouldn't this essentially need to be a high speed camera with very few pixels?

A camera sensor 'pixel' is just a device which conducts proportional to the amount of photons that hit it. A typical digital color camera uses CMOS chips to do this, with a filter on top of them to isolate red, green, and blue. It is pretty basic; the real trick is getting millions of them on a 1/4" sensor and having them relay the data properly with a reasonable amount of noise.

So, yes.

The technology is mostly in those two camps because it's new, expensive, and niche, like WiFi was 20 years ago. As production increases (starts?), components will drop in cost and become standardized and more ubiquitious. At the moment there is a lot of practical research you can read about from conferences like OFC [1] and SPIE PW at the free-space laser or telecom tracks [2]. At the moment transmitters and receivers are made up of highly specified components for their use case and are very parametrized, for example there are hundreds of different DFB lasers that can be used as sources.

I'm a little surprised that IEEE has already standardized, especially given the wavelength they chose but I imagine their members were forced to adopt a prolific technology as without a standard they risk the technology moving ahead without them.

[1] https://www.ofcconference.org/en-us/home/about/archive/ [2] https://spie.org/Publications/Proceedings/Volume/12413?&orig...

Basically, the only new principle involved is that instead of de-serialized and subcarrier modulated data modulated onto 2.4GHz sinewave and emitted from antennae as electrical field changes, it is now sent as changes in light level on subcarrier frequency.

Or more simply, maybe it could be done in YouTuber style by a light-emitting diode on Tx antenna port and a photo-sensitive diode on Rx antenna port? Switching speed of Tx side LED could become the limiting factor in that case.

Also to a more trivial extent, perhaps the same idea behind the IR ports on Gameboy Colors. Which is a really interesting and probably underutilized feature to think back on now. And even more interesting in that this preceded the first mass consumer devices offering WiFi, which Wikipedia tells me only really took off with Apple’s iBooks in 1999.

Well I remember syncing my Palm Pilot via IrDA (but it was connected to the serial port, I just had a tower PC)

IR was great, you could do things like transfer contacts and data between palmtops.

I mean I got a palmtop (a Palm V iirc) for cheap well after they were commonly used and I never used it for anything important, but still, it was a cool device. I think I have it somewhere still, wonder if it still works. I mainly used it to play Sudoku on though.

I immediately thought of https://en.wikipedia.org/wiki/RONJA

> My first thought on reading that headline: isn't that single/multimode fiber?

Kind of how radio being a wireless telegram system:

> You see, wire telegraph is a kind of a very, very long cat. You pull his tail in New York and his head is meowing in Los Angeles. Do you understand this? And radio operates exactly the same way: you send signals here, they receive them there. The only difference is that there is no cat.

* https://quoteinvestigator.com/2012/02/24/telegraph-cat/

This time the tail is fibre and not copper, but without the cat tail.

WiGig just came too early, the underlying tech was also barely able to support it. Windows 7 and Vista era, expensive and the software really wasn't there either.

It might be different now with VR and the rise of docks.

The wireless adapter for the HTC Vive headset was based on WiGig. The range is limited to a few meters, but that's more than enough for room-scale VR. The need for massive bandwidth and low latency make it a perfect use case for WiGig.

Wigig wasn't fast enough to be useful for wireless HDMI, especially given 4k displays arrived right at that moment, and 2nd gen WiGig it's probably way to expensive.

The question is whether it's possible to create a full-rate 224 gbps transciever that is cheap enough to appear in mid range phones, notebooks and TVs and whether it would work without line of sight, but in the same room, with reflected light.

Clicked into the comments hoping such a comment would be here.

Of note is that the first commercially available Li-Fi system has been available since 2014, and it hasn't gained any traction?

reading that, it seems to be about consumption/receiving data rather than the sending of data. or do you have light shining out of your laptop back at it?

there is an example of a school using it, and the students have a usb device in their laptop to read the light, but is it slow upload or are they also connected to wi-fi?

"LiFi" does seem pretty niche. Anywhere you're okay with a 10-15m range, have full line-of-sight, somehow can't run a wire or fiber run, and don't want regular WiFi for whatever reason.

Niche: video editing, interacting with AI/ML training clusters, ...?

high fidelity AR/VR without cabling

iPhones that still have USB 2.0 ports /s

The use case can be found here: https://lifi.co/lifi-applications/

I can only think of useful for conference where everywhere there's light.

Wireless hdmi, 4k webcams would be nice.

10x 22.4GB/s download from CDN or 1x 112GB/s Li-Fi to Li-Fi local transfer before inevitable massive derating

Synchronizing LLMs, streaming AR textures, downloading Linux ISOs. If you give someone bandwidth, they will fill it.

For you home network, they're are likely few or no use card to warrant a new tech like this.

But for offices, schools, conference halls, and other place with many people together it could be useful as a very fast zero-wire solution. Even then it'll be niche: most office workers or conference attendees don't need 100Mbit let alone more than 1Gbit. But some will: maybe said office is full of video editors and such.

Remember that the bandwidth is a shared resource like WiFi: that isn't "up to 224Gbit/s for every device in the room" it is "up to 224Gbit/s for every device in total". And that "up to" value is for ideal circumstances: there will be some environmental interference (though LOS limitations will work in this techs favour there) and, more significantly, the more devices you have in a given collision domain the more they will interfere slightly with each other too because they can't share the group resource with complete efficiency (and this difficulty grows exponentially after a certain threshold). Get a couple of machines transferring arbitrary data as far as they can via WiFi, take the total of transfer speeds they are getting and now add more machines doing the same. The total might scale well for an irritation or two of this test, but there will be a pint when it significantly won't and the total bandwidth will actually fall (and latency will shoot up).

The main push for 5G mobile networks wasn't better max bandwidth per device, but better performance (latency and throughout) for every device when you have a lot of them sharing the same collision domain. The headline figure of 224G bit is (while not at all dishonest) just that: a figure to gain interest via headlines. Those who actually have a need or want for such tech will be looking much deeper into it than that.

It's "Data transmission tests have reached speeds of up to 224 Gbps" which might translate to something like 50 Gbps, or 6 GBytes/s of reliable goodput. Divide by the number of users. Wireless high res VR/AR gear might indeed be one application but also many normal "access my data over the network" applications on the laptop currently need need this class of bandwidth. Note that on normal ethernet you get the 40-100 Gbit/s BW per end user, since it's wired & switched. But this will be shared media.

I think the applications would have to be in data centers where environments are controlled and certain hardware can easily eclipse 224GB/s such as GPU memory.

But I'd assume that if you're in a datacenter, you can use physical wires since you control everything.

Even if your devices can't produce or consume data nearly that fast, it's still good to have all that bandwidth at the link level because you'll probably be sharing it with multiple devices.

Maybe this would be in some sort of specialised network appliance, bridging the link into a wired network of some kind, not a general-purpose machine buffering the network data in regular RAM?

From rough memory, network switches often have total switch bandwidth figures measured in terabits per second.

So "where several other devices intersect" seems like reasonable first thought for where large speeds are needed.

Modern GPUs routinely exceed 5 Tbps in memory bandwidth. (GDDR6X at 20 Gbps per pin, times 256-bit bus width is 5120 Gbps.)

The original article unfortunately quotes incorrect units: 224 Gbps (gigabits per second) is NOT the same as 224 GBps (gigabytes per second).

> Isn't this require a "clear line of sight?"

It's a very nice feature. Imagine not having to share bandwidth with your neighbors.

> Wi-fi works across walls, around corners and so on. Is this possible with this li-fi?

Think of this as a wireless ethernet cable. Instead of having a jack in the wall, you have a light fixture in the room.

The line of sight does not need to be direct, light can bounce and still retain enough brightness to communicate. Which is also how 5ghz wifi works because 5ghz wifi also has trouble with going through objects like walls.

Not sure about direct line of sight but the article mentions that it'll never fully replace wifi for the reason you've stated. Radio waves can go much further distances and penetrate walls/buildings etc. Light can't do that.

Presumably it will be encrypted.

Are you asking if direct line of sight is needed? I’m wondering as well

In high security environments windowless room are specified. Like a SCIF.

An enhancement doesn't imply perfection. For instance, this would make wardriving far more difficult.

I’m curious to see what VR/AR applications this might have.

Imagine peripheral devices capable of much more than they are today merely because of more bandwidth.

Imagine how useful it would be to have a wireless NAS capable of acting as an iSCSI target without having to worry about running cables.

And as “spatial” computing use cases increase along with extremely high res formats like ProRes, we’ll quickly find ways to use the extra bandwidth.

The same reason end users want more than 640k of memory.

It probably doesn't, except for a few cases like a video editing station not working with entirely local resources.

But when you have a lot of devices in the same collision domain (like WiFi, and hubbed rather than switched networks of old) this is a shared network leg: get a 50+ devices in the same room pushing data back & forth at various rates and I doubt you'd see anything like the theoretical total bandwidth of 224Gbit/sec for that network leg (and latency will shoot up). Maybe you'll still reliably get 1Gbit/device which you wouldn't with other tech, but you won't see anything like 224/50+Gbit per device. Try doing anything much on shared WiFi at a large conference, or on your phone's non-wifi data capabilities in a large public venue with hundreds of other phones even just idly interacting with the network, and you'll see what I mean.

The speed figure is just a headline figure, and attention grabber, not actually misleading but not at all as meaningful as the headline writers might think. The more important details are how well the tech works when congested, how much how many devices can do before the effective available bandwidth falls through the floor and/or latency figures reach for the moon. That 224Git figure is the upper limit for the whole collision domain (the room, as this is an LoS constrained technology), much like 56Mbit was the upper limit on 802.11g and 11Mbit was the upper limit on 802.11b (how often did you see those rates even on aggregate with more than a couple of devices active at a time?).

I believe we're talking about the equivalent of a small LED, like half your devices already have one or more of, but infrared and probably lower intensity. I.e. a teeny invisible glow. Even for animals that can see the glow, it would low enough intensity to be barely noticable.

If it is used as a light bulb, I am sure someone will be able to see it. Some light bulbs flicker. And it causes headache.