Stanford researchers develop full-duplex wireless

 

Three antennas to cancel noise.

Stanford researchers have developed a prototype full-duplex wireless node capable of simultaneous transmissions on the same frequency, which they claim could double wireless network speeds.

Today, wireless networks are "half duplex", meaning that they can transmit information both ways, but not simultaneously. Devices, such as two-way radios, highlight that limitation.

Full duplex landlines and Ethernet connections, on the other hand, make use of two separate cables to send and receive transmissions simultaneously.

One of the main problems that Stanford electrical engineering researchers Jung Il Choi, Mayank Jain and Kannan Srinivasan identified with wireless radio was that the noise caused by a radio's own transmitter prevented the radio from receiving and sending transmissions simultaneously.

"When a radio is transmitting, its own transmission is millions, billions of times stronger than anything else it might hear [from another radio]," explained Philip Levis, a Stanford assistant professor of computer science who helped the researchers.

"It's trying to hear a whisper while you yourself are shouting."

The researchers' key innovation was Antenna Cancellation, which they built into a radio receiver to enable it to filter out the signal from its own transmitter, leaving it capable of receiving weaker signals from an external radio.

"The challenge is canceling the node's own transmitted signal (shout) from what it receives (whisper)," they explained on their design page.

The node relies on three antennas rather than the two found in conventional transmitters to create a process similar to noise-canceling headphones.

The device exploits the fact that each radio knows what it is transmitting and therefore knows what its receiver should filter out.

"Our design uses two transmit antennas [and] one receive antenna per node," the researchers explained.

"The transmit antennas send the same data and the receive antenna is placed such that there is destructive interference from the two transmit antennas, thus reducing self-interference.

"Offsetting the two transmit signals by half of the wavelength causes them to cancel each other, creating a null position where the transmitted signal is much, much weaker."

The researchers noted that some obstacles still exist, including a "fundamental limit on bandwidth" and power constraints caused by the cancellation technique.

"The use of narrowband, low-power, active noise cancellation devices further exacerbates these restrictions. We are working to improve the design of full-duplex radios to remove these limitations," the researchers said.

Copyright © iTnews.com.au . All rights reserved.


Stanford researchers develop full-duplex wireless
"@MS, well let's have a look at a couple of those points: ""@Maxxi2 "Great addition to Fibre infrastructure, as wireless has always been" Not sure which country you are referring to."" Do some ..."
By Maxxi2
 
 
 
Comments: 13
MerariSchroeder
Feb 23, 2011 4:40 PM
Fantastic. A good example of how spectral efficiency is not the only front in wireless research. I had always assumed full-duplex was not possible with wireless, because of the waves cancelling each other out in transit. Turns out it was just a "big mouth" and little ears. Full duplex will mean much better latency in the future and the ability to eliminate an entire upload channel for mobile internet. Does Conroy and NBNCo still think wireless is a dead end? Do they still ignorantly believe that wireless poses no threat to their FTTH-only business model? Wireless is a real threat - http://nbnoptions.org/Threats.aspx
mad1k5
Feb 24, 2011 1:08 AM
Rubbish MerariSchroeder,

3 Antenna's you got to be kidding!

Antenna's getting bigger and bigger, As they advance they will require more Antenna's.

Wireless is NOT a THREAT.

They work well together, but not by themselves.
Maxxi2
Feb 24, 2011 7:27 AM
All these wireless technology developments are exciting MerariSchroeder, and complement the FTTH rollout and infrastructure well, ONCE they have been actually productised, proven to be economically viable and have been adopted by the market.

Great addition to Fibre infrastructure, as wireless has always been and will continue to be, once if has been proven to actually function in the field and telcos have been shown to actually be ready to invest in it.

Interesting comment on just a couple os aspects that will take some time to resolve, if and when they can:

"The researchers noted that some obstacles still exist, including a "fundamental limit on bandwidth" and power constraints caused by the cancellation technique."

Nothing that cannot be advanced over time, good news.

In the meantime, I guess you will also note that people just as smart as these guys, and 100 times as many of them, continue to advance FTTH technology capabilities, that the already available 1Gbps systems for FTTP/FTTH will be distributed and used, and higher performance systems will continue to come to market and be used, as they have already been proven to be economically viable.

Fantastic, eh?
Ace
Feb 24, 2011 3:12 PM
Indeed wireless is a fantastic, even expected complement to fixed line broadband. And the stats show exactly that.

People definitely love wireless data on their phones and tablets, and even notebooks PCs for some. The stats show a huge uptake in connections, but bugger all data transfer. Very few people consume or send any quantity of data while mobile. That occurs when they're at their desk, or on a PC at home. It's all there on the ABS web site.

I think it is difficult if not impossible to overcome the problems with wireless, at least in the short to medium term. Like, you can't run it underground and out-of-sight. And it's going to compete over uncontrolled space etc.

@MS, I thought the NBN did have a wireless plan, and that in fact they recently purchased some wireless spectrum. I can't see where this FTTH-only business model is. It's obviously not the NBN's business model. Can you point out where this is spelled out? And I'm not sure what you mean by "wireless is a real threat". Are you worried about getting cancer from a massive increae is radio transmissions? I know a lot of people are - look at the fights over current cell towers. However, I'm a little dubious that there is any real danger to humans from such transmissions. Studies seem to indicate it unlikely to occur.
HubertCumberdale
Feb 24, 2011 3:34 PM
So this is the big breakthrough the wirelesstards were hyping? I'm disappointed, I guess I wrongfully assumed it would be something really special, guess we will be waiting a bit longer for that star trek stuff after all :-(
anonymous
Feb 24, 2011 4:44 PM

@Ace, go easy on MS - it can't be easy to robotically keep insisting that black is white when he doesn't seem too clear about some of the key tech issues involved in his NoNBN campaign.
;-)
Mordd
Feb 24, 2011 8:17 PM
I'm guessing he (MS) goes around and posts once on each website then never returns to the same article to actually discuss/debate with the other commentators....?

I liken ppl like him to scripts that do the same thing, but without all the wasted human effort manually posting everywhere. Which I guess makes them pretty stupid overall.
HubertCumberdale
Feb 24, 2011 9:34 PM
Mordd wrote:
I'm guessing he (MS) goes around and posts once on each website then never returns to the same article to actually discuss/debate with the other commentators....?

It happens, just very rarely. I think mentioning NoNBN.org has something to do with it.
MerariSchroeder
Feb 25, 2011 9:29 AM
@HubertCumberdale
"posts once on each website then never returns"
Generally. I don't have time to revisit all articles that I commented on.

If you want to have an efficient debate, do so on my website, there's plenty to talk about - http://nbnoptions.org/Balance.aspx

[mad1k5] "Antenna's getting bigger and bigger"
There is 60Ghz technology where the antenna is packaged into an MEMs IC (by NICTA). To me that's smaller and cheaper.

"Wireless is NOT a THREAT.

They work well together, but not by themselves."
Wireless can work without Fibre. Especially with the provision of 10Gbps directional backhaul.

@Maxxi2 "Great addition to Fibre infrastructure, as wireless has always been"
Not sure which country you are referring to.

"..continue to advance FTTH technology capabilities"
If the market is well serviced by wireless, why would they bother duplicating their service? I fail to see any reason that a residential customer needs more than 100Mbps. What content stream requires more than that? Pictures? Sound? Video? 3D Video ? .. ? And what is so compelling, that requires all of Australia to pay more for it, when for the history of the internet it has always decreased in price?

@Ace "I think it is difficult if not impossible to overcome the problems with wireless"
Too many people talk with experience of current technology. Or experiences with Optus when the iPhone came out or Vodafone (I was with them, couldn't even get into my internet banking while in the city). I'm now with Telstra and there is no perceivable issue - I even RDP with no perceivable delay, and I will be more than happy when mobile services are increased 20x with LTE-A services.

You can't say that 20x (42Mbps theoretical to 1Gbps theoretical) won't make a difference. I guess the narrow minded can and will keep trying to implicate contention, when it clearly works today (not forgetting that there is also be contention with FTTH).

I just ran a speed test on my phone. It was .5Mbps. That's
without any ability in my phone to use the MIMO abilities of Telstras 42Mbps theoretical speeds. So it's effectively 21Mbps. So multiply that by 40x and you get 20Mbps - pretty much the speed you'll get with the lowest end plan (and the only affordable plan) from NBNCo. So what will customers do? Pay for both their mobile plan (which already includes internet) and a fixed line NBNCo rip off special? I think not. (Of course I exclude those NBN supporters here who clearly have much more money than the 16% of Australians who earn less than $400/wk and won't be able to afford a shiny fibre internet connection).
HubertCumberdale
Feb 25, 2011 11:12 AM
MerariSchroeder wrote:
Generally. I don't have time to revisit all articles that I commented on.

Of course not, you're too busy raising $50+ billion to stop the NBN.

MerariSchroeder wrote:
If you want to have an efficient debate, do so on my website, there's plenty to talk about - http://nbnoptions.org/Balance.aspx

LOL x1000
ray73864
Feb 25, 2011 12:16 PM
@MS
Wireless can work without Fibre. Especially with the provision of 10Gbps directional backhaul.

Where is that 10Gbps backhaul going to come from? I assume we will either have really really big fat pipes, or a few little tiny fibre optic cables.

Either way, if we moved to an all wireless broadband network, we would still need to lay down fibre optic cable in order to provide the necessary backhaul for all the necessary wireless towers that would have to be placed.
Ace
Feb 25, 2011 1:03 PM
@MS said: "Too many people talk with experience of current technology. Or experiences with Optus when the iPhone came out or Vodafone (I was with them, couldn't even get into my internet banking while in the city). I'm now with Telstra and there is no perceivable issue - I even RDP with no perceivable delay, and I will be more than happy when mobile services are increased 20x with LTE-A services."

Sorry, I hadn't quite understood your argument previously. I now understand that if it works OK on your phone, then it's an acceptable technology that'll work for everyone. Got it.
Maxxi2
Feb 25, 2011 2:24 PM
@MS, well let's have a look at a couple of those points:

""@Maxxi2 "Great addition to Fibre infrastructure, as wireless has always been"
Not sure which country you are referring to.""

Do some research into international markets and I am sure you will find some results yourself. There are enough of them.

"" "..continue to advance FTTH technology capabilities"
If the market is well serviced by wireless, why would they bother duplicating their service?""

Because there is no single comparable to our's market globally that is well enough serviced by WTTH, and as such the great wireles services today and in the past are only those that complement the ever advancing wired infrastructure that exists, and will continue to be developed as FTTH/FTTP expands...

""I fail to see any reason that a residential customer needs more than 100Mbps.""

That may be exactly your restriction MS, your failure. You cannot identify these reasons. But then again we heard the same statements from industry folks when we moved from 9.6Kbps to 14.4Kbps to 28.8Kbps to 56Kbps to 64Kbps ISDN 128Kbps ISDN to 256Kbps DSL to 512Kbps DSL to 1Mbps ADSL and ever onwards to the 24Mbps ADSL we can connect to in many premises, businesses, servcies providers et al.

Sorry MS, but a tired onld and broken record on that one my good fellow.

There is zero (and spell that in as many languages as you like globally) reason, evidence or statistics that support the assertion that the growth in high bandwidth enabled services, commerce, communiations and general internet usage has all of a sudden stopped in 2010...

>;))


""What content stream requires more than that? Pictures? Sound? Video? 3D Video ? .. ? ""

Sadly it would seem that you either do not know what the open networking infrastructure is used for today (ie what the Internet is used for), or you desire to avoid discussing the services growth that is enabled and enhanced by higher bandwidth availability.

This may be the reason that you seem to either completely miss the point on this question, always, or you are avoiding it.

Do some research on online services growth and developments, but you will need to be ready to accept the facts that you find.

Many of these will be enhanced by dynamic wireless sysyems complementing the fibre infrastructure provided performance and bandwidth availability, and the 7% the fibre does not reach will need alternatives such as wireless.

""And what is so compelling, that requires all of Australia to pay more for it, when for the history of the internet it has always decreased in price?""

There is not a single shred of evidence, and by evidence I do not refer to myriad statements from NBN and FTTP opponents or entrenched critics, that NBN profided FTTP connectivity will cost more per unit of service that folks pay today.

By evidence, btw, I mean cold, hard, incontavertible and proven facts. Predictions based on comparing 50Mbps Telstra connections FTTP to 12Mbps ADSL connections today from TGP are not comparisons, they are misleading red herrings...

BTW, the average spend per person on Internet services today is actually higher than it has been in the past MS, it is the cost per unit of service delivered that has come down, and will continue to come down for sime fundamental reasons that you do not seem to comprehend:

1. I can have and pay for a 20Mbps ADSL connection today and get 8Mbps delivered to my premises and have around 1Mbps upstream. This is copper, network age, distance and contention rates conditional. This is not unusual. Let's say I have a median $60pcm package.

2. With FTTP and the same premises and 20Mbps connection, I will get a minimum 12Mbps delivered, with empical evidence showing that it will be closer to 15Mbps with contention rates affecting a %% of the 20Mbgps.

Thus I will be paying less for my services under a FTTP NBN MS. That has been and will continue to be the case, as volumes across an industry have and will continue to bring the unit costs down.
Comments have been disabled for this article.
 
 
 
Top Stories
Australia turns to homegrown drones
Debating the finer points of unmanned aerial vehicle design.
 
The New Zealand telco problem
Opinion: Could Telstra save Kiwi telcos?
 
IT price probe to 'name and shame' gougers
Industry ducking the issue, committee claims.
 
Sign up to receive iTnews email bulletins
   FOLLOW US...

Latest VideosSee all videos »

Latest Comments
Polls
Should the Government enact new legislation to protect copyright holders in the digital age?

   |   View results
Yes
  19%
 
No
  81%
TOTAL VOTES: 511

Vote