Reflections

RevMike writes "The New York Times (reg required) is reporting that Bell Labs/Lucent has developed a method to multiply the bandwidth of cellular networks by using multipath. Robert Lucky developed the system called Blast. He claims that it should multiply the data rates in existing spectrum 300% to 400%. One prototype took a network from 2.5Mb/s to 19.2 Mb/s. Interestingly, the system works better in cluttered environments." And on a related note, Kimberley Burchett writes "The latest Physics News Update mentions that skyscrapers could actually help wireless communication. 'The more scatterers between the transmitters and receivers, the more channels that are available. For the time being, the communication technique is limited to ultrasonic communication - the electronics necessary for exploiting scatterers with wide-band time-reversal antennas at cell phone frequencies simply don't yet exist.'"

Comments

time reversal antennas?

[random735]

is it just me or does it seem unlikely that this
technology will *ever* exist? :]

Re:Blast?

[s20451]

Well, this one appeared quite recently. It turns out that sparrows don't like to live near cellphone towers, and that fact is being blamed for a large reduction in the British sparrow population.

At the present time, the scientists who believe that cell phone radiation causes health effects in humans are a small minority. However, everyone agrees that cell phone use can be hazardous, for example, as a distraction while driving.

mixed blessing?

[tps12]

It's great to hear that technology is improving. I almost dread getting calls from my friends with cellular phones, because the sound quality is so bad and they frequently get disconnected.

But I'm a little skittish about jumping into this hole, hog. We still haven't seen any unbiased studies on the effects of cellular phones on the brain (where people hold their phones while talking) and the reproductive organs (where they keep them when they're not).

I don't know if we want to double, treble, or quadruple their radiation emissions before we know what the effects on living tissue will be. What government department does this sort of thing fall under? The FCC? If there isn't one, then one should be created. Just my two cents.

This isn't exactly scalable.

[Chmarr]

What they're proposing is a method of using channels from adjoining (or not so adjoining) cells to improve bandwidth, making the assumption that the adjoining cells has bandwidth to spare. Now, if you're having trouble getting a call through on your EXISTING cell (or data connection, or whatever), what makes you think that stealing bandwidth from adjoining, likely equally-congested, cells is going to help?

Cell Phones = Cancer is BULLSHIT

[siskbc]

Does anyone have any updated statistics on cellular safety? I wonder if this technology will affect that aspect of cellular use?

OK, I'm a physical chemist, and this junk drives me up the wall.

All these journalists assume that the wattage is what matters, and that being 1 cm from a 4 watt transmitter could be unhealthy. However, the wattage is irrelevant - Einstein showed this with the photoelectric effect. Basically, it is the frequency of the radiation that is dangerous, not the wattage, because one photon interacts with one electron - so the number of photons is irrelevant. For example, which would you rather be near - a 100W light bulb, or a 1W gamma ray emitter? Thought so...;)

To give you a baseline, cancer due to radiation occurs because photons of succicient energy actually break chemical bonds in your DNA. This requires low-UV or better to accomplish. That's why sunburn gives you cancer. Now, compare this to radio waves, which are far, FAR to low in energy to accomplish anything of the sort. In fact, radio waves are even to low to excite the vibrational or rotational states of a molecule (which is how a microwave oven works), so there is no risk of "cooking" your brain, either.

Ultimately, when pressed, these cell-phone-cancer freaks point at two cases where some habitual cell-phone user got cancer on the right side of his brain. Ergo, it was the cell phone. These "doctors" (and I use the term loosely) have never proposed any sort of mechanism or ANYTHING to explain how it could occur. Because it can't.

Bottom line, you have a better chance of getting cancer from your own body heat (you emit infrared radiation) than you do from a cell phone.

And you wanted to know WHY we need faster chips?

[MystikPhish]

"One restraint on speed is the intense data processing it requires. With current technology, higher speeds would demand chips that are too large and too power-hungry for hand-held devices."

So is this the next killer app to fund chip research?

HOW THIS WORKS

[goombah99]

Here's how I imagine that the physics of this works and I'll explain what a time reversal anatenna is.

First how is it possible to gain bandwith using 'clutter' ? here is how. imagine that you and I are both in our cars at a stop light and listening to the same rasio station. Only my radio can barely get the staion and yours does fine. If I pull my car forward a few feet, suddenly my reception improves. What happened is that the multi-path interference made my car sit in a node and your sit at a maximum.

Now if there had been more than one radio transmitter and some dude back at the station had changed the relative phase of the antennas he could have put your car in a node and my car in maximum. Thus by changing the relative phases of antenna one can direct the signal to whichever car you want. If we now differenlty modulate the signals on the antenna's we could send one signal to one car and the other signal to another car. Thus by recycling the same broadcast channel we have doubled our bandwidth.

But there is a problem here. How does the guy at the station know how to adjust the phases of the transmitters to get this spatial separation? the answer is he cant know unless he knows where both of us are and where all of the building are, and the moving cars,etc... basically impossible to compute beforehand.

So instead what you do is have the car transmit a signal on the same frequency that is coded to say 'hi i'm car #1'. The guy at the station receives this signal on his antennas and notes the relative phase of the singals on each. Now by time reversal symmetry, if he broadcast at those phases all of the multi-path signals would converge in phase on my car's antenna. Likewise he can pick out car number 2 and so on.

Furthermore i'f the clutter is changing or almost equivalently if I am driving my car, he can just keep updating the relative phases and track my car's antenna.

So the missing ingredient here is some way to detect phases and re-transmit phases in real time. One approach is to have a reference signal all of the transmitters are locked to that from which they could compute the phases they need to re-transmit. This is potentially compuationally expensive since we also have to demodulate and detect all the signals as well.

another approach is to simply phase conjugate the incoming signal, amplify it, remodulate it witht he signal, and rebroadcast it. Thus the outgoing signal is the time reversed image of the incoming signal. We never need to actually measue the phase.

there are lots of ways of 'phase conjugating' as signal but I'm not enough of a microwave jock to say how you do it in that region of the spectrum. In the optical band region there are lots of ways using non-linear optics. However none of these are wide band. They only work at specific (laser) wavelengths that can be created coherently. From the comments in the article I am assuming this is true in the microwave regime as well. When you get down to the ultra-sonic regime you get to frequencies (mega hertz) where you can do this electronically directly. So that is proably why its accoustic.

interestingly there are a number of approaches to making passive approximate phase conjugate mirrors using engineered materials (bulk element transmission lines) that do operate in the microwave regime. these however are not advanced enough for practical use yet. currenly they are finding use as light weight stealth materials aimed at radar invisibility. But probably within 5 to ten years these will be practical enough for the applications envisioned here.

I might specualte that just as the development of electronics was first spurred by military use and then by consumer use. This might happen here too.

and if some bozo complains about my typing skills I will adjust the phases of my heat ray to melt their brain.

Re:HOW THIS WORKS

[wowbagger]

You were close, but not on.

BLAST doesn't require the receivers to send anything back to the transmitter.

Instead, the receivers use several antennas to create a virtual antenna with great directional sensitivity. Then the receivers use the different reflections of the signal of interest to build up a signal to receive.

This gets tricky to describe without drawing on the wall, so bear with me.

Consider a receiver with 4 antennas at different locations - for the sake of discussion say a square 1 meter on a side. We will also assume 2 transmitters at different locations. The transmitters are on the same carrier frequency, but are transmitting different signals.

For the receiver, there will exist 4 signals (one for each antenna) from each transmitter, for a total of 8 signals.

The receiver creates a set of 4 phase shift networks, and sums their output.

For each transmitter, there will exist a setting of the 4 phase shift networks that will bring all the signals from that transmitter into phase, causing them to reinforce in the summing node. Since the other transmitter will not have the same phase relationship, its signals will not tend to add up. This gives the signal of interest more level than the signal not of interest.

Now, add in some buildings, mountains, lakes, and so on, to create multipath signals. Now the number of signals from each transmitter has increased manyfold. The more signals, the more attenuation the transmitter you are trying to ignore vs. the transmitter you are trying to receive.

However, to accomplish this you need to be able to shift the phase of your network by up to tens of milliseconds, with a resolution of tens of nanoseconds. If you are moving, you have to constantly evaluate how the received signal parameters are changing, and adjust your network accordingly.

And that is why this hasn't been done for RF. For ultrasound, you are looking at needing to delay seconds, with sub-millisecond resolution.

However, when CDMA was first conceived of, the hardware to implement it would have taken several tractor-trailers to carry around. Now it fits in your back pocket.

Robert Lucky and old news

[Doctor K]

First and foremost, this was developed at Bell Labs. (People I knew when I was a Bell Labs were doing on the research.)

Robert Lucky was at Telcordia, not Bell Labs. The New Yorks Times articles notes this. So, editors, please do a vague glance in the general direction of the article and the article summary before posting..

Second, this is old news. Here is a general scientific article on the underlying basis for the technology from September 2001:

http://www.aip.org/pt/vol-54/iss-9/p38.html

Kevin