SoylentNews is people
SoylentNews
Tech Review reports an "impossible" development, https://www.technologyreview.com/lists/innovators-under-35/2016/inventor/dinesh-bharadia/
Because the signal from broadcasting a radio transmission can be 100 billion times louder than the receiving one, it was always assumed that outgoing signals would invariably drown out incoming ones. That's why radios typically send and receive on different frequencies or rapidly alternate between transmitting and receiving. "Even textbooks kind of assumed it was impossible," Bharadia says.
Bharadia developed hardware and software that selectively cancel the far louder outgoing transmission so that a radio can decipher the incoming message. The creation of the first full-duplex radio, which eventually could be incorporated into cell phones, should effectively double available wireless bandwidth by simply using it twice.
Any bets on when this will make it to production, maybe as part of 7G(eneration) wireless? Or will the technology go black, used first by military?
And, does a person's name ever influence their career? "Bharadia" sounds awfully close to "bi-radio"...
(Score: 2) by Immerman on Thursday September 22 2016, @09:54PM
Just *how* good the engineering is.
The problem with echo cancellation is that it's not perfect. But that's okay, because the echo is generally smaller, or at least not much larger, than the original signal. So if you can eliminate the echo with even 90% accuracy, you're left with random noise that's less than 10% of the amplitude of the signal you're trying to receive, a 10:1 signal-to-noise-ratio (SNR). You can usually work with that.
In this case though, you're dealing with an "echo" that's millions or billions of times larger than the incoming signal. It has very little lag, and comparatively low uncertainty, but even if you can deliver really incredibly impressive, near-perfect signal cancellation, eliminating 99.999% of the transmitted signal, you're still left with noise that's 0.001% of a million times bigger than the incoming signal: 10x as much "random" noise as you have signal. That's going to be like listening for ghost voices in TV static.
And these folks are claiming they can reconstruct the received signal despite an "echo" a 100 billion times larger than the incoming signal. To do that, with the same "adequate" 10:1 SNR as in the first example, they need to eliminate 99.9999999999% of the echo. That's not just good, that's practically godlike.
(Score: 2) by Immerman on Thursday September 22 2016, @10:09PM
I realize a bunch of 9s doesn't necessarily mean much to a lot of people, so let me give a more concrete example:
The moon is 238,900 miles away. Being able to eliminate that much noise is akin to being able to map, from here, the height of the moons surface with 1/8" accuracy
Or alternately, to be able to stand in LA, and measure the thickness of a single strand of hair in New York
(Score: 2) by Geotti on Friday September 23 2016, @03:33AM
And for those MBAs you could just say it's 12σ, that's 2 times more 9's than with 6σ [wikipedia.org]!
(Score: 0) by Anonymous Coward on Friday September 23 2016, @12:16PM
Except it's not.
(Score: 2) by Geotti on Friday September 23 2016, @09:50PM
I was referring to the amount of nines.