Slashdot Mirror
Improving Wireless Networks
FOE writes "Picked up this story from Eurekalert. Describes disco-ball like reflector to help IR networks. I really like the name: "chaos mirror" (grin). " Fairly straight-forward device: Takes the incoming beam and spreads it out over a wider range - but it's all IrDA, which has terrible range. I'll keep my ZoomAir, thanks.
rather interesting.... definately has applications... sit down in a room w/ a "disco ball", crank up your laptop and let the networkin' fly!@
.
Could you use one of these to sort of fire off a big burst of laser into the room, then intercept them all as they bounce back and get a kind of image of the room? This could be really useful in robot navigation. Let the robot map the entire room at once.
Then again, maybe they already do that...
Kintanon
Check out JoshJitsu.info for Brazilian Ji
Get the smoke machine fired up and have a disco in your room as the computers and appliances fire information around the network.
So do you have to have one of these boxes for every computer you want to hook up? So basically, the light gets dispersed into a million directions. Does a receptor only need to capture a fragment of the Ir beam?
"Evil will always triumph over good, because good is dumb." - Dark Helmet (Spaceballs)
Why the f**k do you sad and lonely people race to see who can be the first to post a reply? Well now i am starting a new race - who can be the first to slap the sad twat that does the first post! slap
Hey, you could use that to bulid a wireless beowolf cluster
Line-of-sight technology has no business in this increasingly-networked world. This disco ball idea is nothing more than a kludge to try to work around IR's primary limitation.
= -=-=-=-=-=-=-=-=-
Sure, IR was great when all you needed it to do was change channels on your TV set. But now you have two or three TV sets with VCRs and stereo components and every damn one (except Bose) uses the same crappy line-of-sight IR technology.
This idea, honestly, is nothing new. I'm only sorry I didn't patent it. For years I've been using mylar loops to accomplish the same results. Just goto a novelty supply store and pick up a roll of mylar film (its basically plastic with a mirror surface). Cut strips of this stuff and put it in corners around your house. Also try to put mylar on any surfaces surrounding the IR receiver.
By aiming my IR remote at the right corner, I can actually control my VCR or cable box from other rooms and even down a flight of stairs. But it's still a kludge.
I wish the FCC would open up more of the lower ranges for consumer use so that cheap RF remotes could finally be implemented. Apple, get your AirPort to work with a PCI or PC Card interface, puh-leeze!
Remember...IR is (in the most basic sense) heat...so IR networking is just a NIC having hot flashes. The technology is tired and I sure hope that it's not still around in ten years.
My $1.00 - $0.98 not necessarily yours...
- JoeShmoe
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
-- I wonder which will go down in history as the bigger failure: the War on Drugs or the War on Filesharing
(if I can ever break him from aiming it at his knee)
Seriously, I've been been bouncing it off of the ceiling light fixture, light colored lampshades, etc, for years.
I see even classic Slashdot is now pretty much unusable on dial up anymore.
Sheesh. When will these people just get a clue and switch to RF?
One part of a system that I'm working on uses a 2 megabit RF network. It works well, the parts are all fairly cheap. The only real problem is that of power. Since these are PCMCIA cards mostly in laptops or handheld devices, they frequently don't have to power to transmit over a great distance or over a lot of noise. But fixing this is easy. You see, somewhere you have to have a central station that hooks into your wire network. So, you just massively overcover the area with these stations. They're small and cheap too. And practically no setup, although they can be administered remotely by telnetting to one of them.
It's beautiful. Total of maybe 100-120 RF devices in the plant, and 250-300 base stations to get total coverage of every place you just might happen to be. Note that this is a manufacturing plant with one whole hell of a lot of RF noise produced by the machinery, so this overcoverage is necessary. In my office, my laptop (along with about 30 other peoples) goes through one base station at 2 megabits, from anywhere in the building. Only problem is when the wire network goes down, or someone unplugs the base unit. In the plant, of course, the base stations are mounted high in the rafters, so you can't see them anyway and your workers aren't doing something stupid like unplugging them to plug in the coffee maker (which happened once). Plus, adding any new devices to the network (stationary or not) is easy. Just plug in a radio card.
Anyway, RF devices are not new. These cards have been around for a few years, and work fairly well with a little forethought and planning. IR is a clumsy hack whose only use is close range connections. Wiring an office or anything like that is just ugly. MIRRORS to transmit your data around the office? You must be insane.
---
- Give a man a fire and he's warm for a day, but set him on fire and he's warm for the rest of his life.
I agree with the recent followups....IR is *not* the way to go. I'm much more interested in Bluetooth but I don't see much activity for linux.
If you stack four Christmas balls in a pyramid (like cannonballs) you can "see the Chaos" inside them in the form of a fractal basin boundary.
This boundary has the "Wada property" which makes the IR beam spread throughout the room. To demonstrate: (1) take a laser pointer and sight over the top of it to aim the beam at the fractal: the beam will be visible through all of the openings (2) aim so that you miss the fractal pattern but still hit the balls and the beam on exits through only one opening.
For more see: http://www.chaos.umd.edu/~dsweet/Spheres
Dave
The more practical application is to place the sphere in a high location and then use it to direct laser beams onto people's foreheads. They'll never know where you're coming from.
Seriously, this is terribly limited by the 1) line-of-sight requirement and 2) weak IR beams. I'm want my own communications satellite.
I made an optical networking hub a couple of years ago. OK, so it was just a drinking glass... but it did enable 4 IR-equipped Powerbooks to see each other all at once.
IR sucks for a lot of things, but give it a break. I don't want RF in my Palm III... IR is just fine for zapping notes and apps to other people.
The problem with Radio Wave technology is that it is even longer wavelength than Infrared in the electromagnetic spectrum. Order of wavelength from longest to shortest is.
Radio Waves
Microwaves
Infra Red
The Visible Spectrum
Ultra Violet
X-Rays
Gamma Rays
The Longer the wavelength the less directional the signal is. If you are using a point source radio station the signal propogates (and pollutes) that bandwidth 360 degrees in every direction. The signal sort of expands like the surface area of a balloon. This is why radio signals weaken quite rapidly with the square of the distance from the source.
I agree infrared isn't much better. Once you start using multiple sources and receivers in the same room the long wavelengths start interfering with each other rapidly. Still you can focus infrared into a crude wide angle beam usefull for some applications.
Lasers and Optical fiber are almost ideal carriers since the signal loss over distance is much less.
Laser beam signals weaken by the square of the distance as well but, Since a laser beam can be initially focused into a narrow angle beam, the signal maintains strength over a much greater distance.
I have noticed that a lot of people criticize IrDA for not solving all the world problems. I agree that IrDA does not solve all the problems, but in fact it solves some problems that Rf does not.
Infrared is superior in "Point and Shoot" applications. If you are at a conference and want to send your vCard to a person you meet, then you dont want to use Rf, since the machine will then ask you which of the 100 persons in the room you want to send the card to. IrDA does usually not require any addressing. Addressing is the main problem why protocols like IP does not work well (to be nice) in ad-hoc networks. I like to print a document from the HP CapShare or Nokia 9110 without selecting which printer to use. I just point and press one button!
But then IrDA does not work very well when I want to surf the net from my laptop on the bus on my way home from work. Here is BT superior since then I could keep the phone in my rucksack. BT works very well for hidden devices.
AIR will give IrDA 5-10m range and non-line of sight capability (can reflect off walls). VFIR will give IrDA 16Mbps. IrDA will however never work well for hidden devices.
BT has however some problems with their standard, and companies have problems implementing it. They suddenly realized they needed a version 2 of the spec before any devices were using version 1 ;-)
I expect IrDA to survive BT simply because IrDA solves some problems that BT does not. I am however also looking forward to BT since it is a very exciting technology.
We all know that ATM didn't concur the world, and I don't expect any other technology to do it either. Instead I expect devices to support multiple technologies which each has different characteristics like number of buttons you have to press to send a document, cost, bandwidth, range etc.
-- DB