France Telecom Getting Serious About 5G in High Frequencies

Gets ARCEP permission to test. Ted Rappaport at NYU has convinced the industry that high-frequency, millimeter wave wireless can work. But many questions remain. What are the effects of walls, windows, rain, and distance? We have only partial answers.

France Telecom/Orange intends to find out. They just receive approval for a year's worth of testing in the ancient city of Belfort. Much of the testing to date has been in Manhattan, Brooklyn and other highrise areas. Belfort is ideal for testing low-rise and suburban regions.

Read more ...

Full Duplex: DT, SKT, Stanford Guys Say Close

Same frequency, same time, ?double wireless speeds. Today, self-interference makes it impractical to send and receive at the same time on the same frequencies. Testing at Deutsche Telekom and SK Telecom suggests Moore's Law is bringing enough processing power to change that. (pr below.) They are working with Kumu Networks, a spinoff from the lab of Stanford's Philip Levis. Their extraordinary advisory board gives them extra credibility.

Research on Full Duplex is hot.  An article by Sabharwal et. al. lists 141 papers (below.) At Columbia,  Harish Krishnaswamy has developed chips. At a Cambridge Wireless meeting, David Lister of Vodafone concluded, "We can consider the problem of self-interference cancellation as solved. Now is the time to consider system requirements and assess the use cases. There are still major challenges to overcome."

Joel Brand of Kumu is confident. "By the second half of 2016 we could ship full-duplex solutions for infrastructure applications where the requirements are a bit more relaxed than in a mobile phone." WISP backhaul would be a natural niche for this kind of product. 

Some are skeptical about the potential to double performance in the real world.

Read more ...

More Spectrum Than We Need: Sprint Drops Out of Auction

Marty Cooper is right: If we are efficient, "There is no spectrum shortage and there never will be." Showing there's enough spectrum if used efficiently, Marcelo Claure concludes "[Our] spectrum holdings are sufficient to provide its current and future customers." Claure has ~100 MHz in high frequencies, so much he is thinking of using some of it for backhaul. (The FCC should stop that, and other many wasteful practices, in the name of efficiency. But true efficiency is generally ignored.)   

Cooper, the inventor of the cell phone, was joined by Marconi Fellows A.J. Paulraj, John Cioffi and Vint Cerf in projecting technical advances that could raise wireless capacity 50-100 times. The engineers convinced me until the companies spent $45B in the last auction. That's 3-4 times as much as the experts I respect predicted. I don't think anyone well informed is certain of next year's results.

Read more ...

40 Million Comcast Gigabit Homes. Really.

Gigabit (shared) to nearly all of 53M homes and businesses. Comcast is going to upgrade 40% of the U.S. to DOCSIS 3.1, offering a gigabit. Brian's boys are going to start in 2016, probably early, and continue for another year or two. Comcast VP Robert Howald dropped a bombshell. "We're testing it this year. Our intent is to scale it through our footprint through 2016. We want to get it across the footprint very quickly. We're shooting for two years," he said in Mike Dano's Fierce Cable interview. The story was picked up by the Washington Post and a dozen others. Everyone in broadband has known for years gigabit cable was on the way and now the big papers are getting the message.  

"Shared" speeds will be 500+ megabits down 95+% of the time, I predict. That's similar to the 400-700 megabit speeds of AT&T's coming "gigaclear" G.fast fiber to the basement.

Read more ...

England Tops in Euro Medium/fast Broadband

Joe Garner of BT claims Britain is ahead of other big Euro countries in "superfast" broadband. BT deserves credit for offering ~80% of Britain 30 meg or more downstream, a little ahead of Germany and well ahead of France, Spain and Italy. "Superfast" is pure hype. BT's deployment is mostly DSL from their 100,000 street cabinets, typically 30-70 megabits. Unlike Germany, Belgium, Swisscom and Australia, England isn't using vectoring to ~double many connections. Half of the UK is able to get 100 meg or more from cable, as well as nearly 90% of the U.S. ~50 meg would have been superfast in 2007 but is only medium in 2015.

One Garner comment in the Telegraph is offbase. French prices by almost any measurement are considerably lower than Britain. Miles Brignall in the Guardian just wrote "Ripoff Britain: why we pay more for broadband than Europe," with a comparison to France at half the price. Garner's comment, "Britain has the lowest landline, broadband and superfast broadband prices among major economies" is a mistake. The original source, Analysys Mason, had it wrong. It should be corrected by sending a letter to the paper. 

Read more ...

AT&T paid $17/month extra for video (Datapoint article)

Consumers at carriers with fewer than 7M customers wind up paying a high price. At Goldman Sachs, CFO John Stephens suggested major synergies for the DirecTV deal due to “There's about a $17 difference on average between the price the U-verse platform pays for content on an apples-to-apples basis than the DirecTV platform pays. And on 6 million customers, you can get your head around about $100 million a month of expense.” That's an important datapoint to understand the U.S. TV market. Hollywood takes most of the gross income. Seeking Alpha transcript, more below.

Possible implications

Read more ...

Quantenna's Remarkable 10 Gig WiFi - Spectrum Greedy But Works

10 gig requires a remarkable 160 MHz+ but it makes a great headline. With this 8x8 chipset, realistic speed will head to a gigabit and higher, in a room with just the right walls for the MIMO to work. Update Sept 15. Quantenna writes me they expect production quantities in 3-4 months. 

Speeds through a wall go down severely with any WiFi, but throughput even through walls should be excellent. MU-MIMO - dedicating the 8-12 streams 1, 2 or 4 to an individual device with beamforming - should give each of the devices excellent throughput. Quantenna has distributed samples but production quantities are in the future.

If you have no neighbors and completely uncontested WiFi frequencies, you will sometimes be able to measure a nominal 9 or 10 gigabit WiFi from Quantenna and Freescale's new chips. You'll need an ideal location, with just enough interference to bounce the 12 separate signals but not enough to significantly slow anything down. 160 MHz is about the entire bandwidth of the combined AT&T and Verizon LTE networks, far more than will be available to most of us in most places. So the many MU-MIMO connections will have more practical impact for most of us.   In addition, they are squeezing a heck of a lot of data in, going to a 1024 QAM constellation, needing extraordinary analog circuitry and minimal interference.  

Twenty years ago, Arogyaswami Paulraj discovered using more antennas could multiply the throughput of a wireless network. That day it rained on the Stanford campus and he moved his multi-transmitter experiments indoors. Because the transmitters would be close together, he expected poor results. To his surprise, the separate signals came in clearly. He had invented MIMO.

Read more ...

Supersonic DOCSIS: 15 Gigabit Cable 2020, 50-80 Gigabits 2030

Tom Cloonan of Arris is confident of remarkable speeds. (Video and transcript below). At CableLabs' 2015 summer event he predicted DOCSIS would reach 15 gig (shared) in a few years and 50-80 gig a decade later. 15 gigabits would require using 1.7 GHz and higher speeds would require going to 6GHz and more. Jeff Baumgartner at Multichannel calls this "supersonic DOCSIS." 

Read more ...