Israel has a thriving startup scene in general, but in particular it’s been a hotbed for mobile infrastructure startups, which have built off radio technologies developed by the Israeli military. The small country produced numerous broadband radio specialists like Alvarion and Ceragon Networks, but lately the startup scene has shifted its focus from building radios to developing network optimization technologies, producing companies like Intucell Systems, eVolution Networks, Flash Networks, Traffix Systems, Vasona Networks and WeFi.

The world took notice of those developments. Last week, Cisco Systems announced it is buying self-healing networking company Intucell for $475 million, while F5 Networks bought network-signaling outfit Traffix in February for an undisclosed amount. Carriers are also getting in on the action. In 2011, AT&T opened up one of its Foundry innovation centers in Ra’anana, Israel, to work directly with regional startups and developers. The IMA also is collaborating with Orange, Telefónica and Tata DoCoMo on other lab projects.

The Sprint project, however, is specifically focused on LTE. Though Sprint didn’t reveal too many details on the lab setup, it appears to be setting up a 4G test network in Tel Aviv, over which network startups can fine-tune their technology and developers can run their apps. Sprint, however, said it wouldn’t attempt to dictate what the startups do in the lab. Instead, Sprint said it is just providing the tools and network frame, while the participants are free to pursue whatever field they want.

“As a complement to our efforts to unlock innovation in the United States, this Sprint-sponsored LTE acceleration lab will allow startups and developers in Israel valuable access to dramatically speed up their time to market,” Sprint CTO Stephen Bye said in a statement.

Photo courtesy of Shutterstock user Jim Barber

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• New solutions for the evolving mobile network
• Mobile 2012 and beyond
• CES 2012: a recap and analysis

But Sprint had a backup plan. It relaunched Direct Connect last year as an IP service over its 3G CDMA network and even extended basic connectivity to its 2G footprint creating a far bigger coverage area than Nextel ever offered. Apparently that work is paying off, at least in part. Sprint has succeeded in signing up 1 million subscribers to Sprint Direct Connect, the carrier revealed today.

Nextel has been a big drag on Sprint’s subscriber growth for years, but last quarter the rate of defections picked up steam as Sprint announced it would shut down the network for good next June. In the second quarter, 1 million Nextel and Boost customers dumped their iDEN phones, but Sprint and its other prepaid brands managed to reclaim 600,000 of those customers.

We’re almost at the end of the third quarter, and the Nextel exodus has probably only grown larger as Sprint has shut down more iDEN cell sites. At GigaOM’s Mobilize conference this week, Sprint CTO Stephen Bye will be on hand to discuss what Sprint plans to do with that old Nextel spectrum – namely build an LTE network.

Push-to-talk no longer has the caché it once enjoyed – we don’t hear the familiar chirp of Nextel phones emanating from suit pockets and purses like we did back in the mid-aughts. But the service still has a sizable fan base, and other companies are looking to capitalize on Nextel’s demise. FierceWireless reported earlier this week that AT&T is preparing to launch an enhanced version of its own Kodiak Networks-built PTT application in November. Third party developer Wave Connections has also launched a competing service, which has the advantage of working across carriers while running on smartphones.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• 12 tech leaders’ resolutions for 2012
• Updated: Forecast: global mobile subscribers, 2010-2015
• New solutions for the evolving mobile network

The FCC’s plan is part of a larger, and quite controversial, proposal first put forth by the President’s Council of Advisors on Policy and Technology (PCAST) to clear 1000 MHz of airwaves for mobile broadband use. That’s far more than the Obama administration is aiming for in its broadband plan, but PCAST’s recommendations all come with the sharing caveat: instead of booting the feds out of their airwaves, the public and private sectors must find a way to coexist.

FCC Chairman Julius Genachowski

The FCC is chipping away at the recommendation one spectrum band at a time, and it’s choosing to start with the 3.5 GHz band. In his announcement Wednesday, commission chairman Julius Genachowski said he would take formal steps by the end of the year to open the 3.5 GHz band for small cell use. While Genachowski never mentioned the word “sharing” in his statement, Ars Technica confirmed with FCC officials that the plan was to make 3.5 GHz shared between its current government radar use and wireless carriers.

All of the major carriers in the U.S. have talked up small cells, though Sprint is the only one that has laid out a specific deployment timeline. Sprint plans to deploy tens of thousands of the tiny base stations in 2013 and 2014, starting with malls and big public venues and then moving into high-traffic outdoor areas. At GigaOM’s Mobilize conference next week, Sprint CTO Stephen Bye will share more details on Sprint’s small cell plans.

The 3.5 GHz band would be ideal for small cell deployments. By creating a separate band for the tiny little networks, carriers wouldn’t have to worry about interference with their primary macro networks. Rather than try to wrap big tower-based macro cells around small cells using the same frequencies, a dedicated band means micro- and pico-cells could be placed directly under the macro umbrella. Each of those small cells would have the same capacity as their larger counterparts, so adding dozens of small cells beneath a single macro cell would boost its capacity enormously.

Also, higher frequencies don’t propagate as far as lower ones, which is why carriers have always sought to acquire lower band spectrum whenever possible. Since small cells, by definition, have much tinier coverage areas, they’re ideally suited for such high-frequency spectrum.

Don’t expect carriers to fall all over themselves to get this spectrum, though. The operators would rather own their licenses outright than share them with anyone, and the big carrier trade group CTIA has said that spectrum sharing in many cases simply won’t work. There are two different ways to share the airwaves — divvying them up by time or by geography — and according to CTIA, both approaches present problems.

In the case of small cells, carriers need that capacity most in densely populated areas where numerous customers are competing for bandwidth within a limited space. So if the government’s radar facilities are geographically distance — out in the desert, say — then spectrum sharing could work. If they’re at a busy metropolitan airport, that’s another story altogether.

The same goes for timesharing. If the government only uses those frequencies in the wee hours of the night, then carriers could easily turn their small cells off as traffic on the network is at its lowest point. The problem is if the Feds need their airwaves during the peak hours of the day, which is when carriers need them most as well.

We’ll see more details emerge as the FCC starts its deliberations, but even if a spectrum-sharing compromise can be worked out at 3.5 GHz, don’t expect the carriers to embrace this deal. They don’t want to set a spectrum sharing precedent, so they’re likely to lobby the government to clear out as much of this spectrum as possible for their own exclusive use.

Photo courtesy of Shutterstock user Everett Collection

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Consumer privacy in the mobile advertising era
• 2012: Data, spectrum and the race to LTE
• New solutions for the evolving mobile network

Three different carriers. Three completely different approaches to the market. The only thing they have in common — besides being part of the newest wave of mobile virtual network operators (MVNOS) — is that they are questioning long-accepted mobile business models. As mobile evolves from a carrier-dominated, vertically integrated and voice-centric industry into a more inclusive, data-focused one, they’re making the case that our fundamental notions of what a carrier is and what it provides should evolve as well.

That transformation will be a major theme at GigaOM’s Mobilize conference next month in San Francisco. David Morken, CEO of Bandwith.com (Republic’s parent); Scott Bendar, co-founder and CTO of FreedomPop; and Ahmed Khattak, co-founder and CEO of GSM Nation will be on the stage at Mobilize to debate the merits of their new approaches to wireless market.

But the transformation of the industry isn’t just limited to carriers. The switch from voice to data has led every link in the mobile value chain to question assumptions formed when the world used wireline networks for data and mobile networks for voice. Device makers have changed the way handsets are designed. Internet companies are grappling with the fact that their customers are moving away from the PC to small-screen devices, with their limited real estate and more challenging revenue models. And infrastructure makers and carriers are re-architecting the fundamental topologies of their networks.

We’re seeing examples of it all over the industry. Facebook’s IPO was clouded by the revelation that it had no idea how to port its advertising-based revenue model over to mobile phones. A growing number of developers are looking to mobile as their first and sometimes only platform. For a company like Foursquared the smartphone isn’t so much a telephony or a computing platform as it is an extension of its owner’s presence in the world. Path (whose CEO Dave Morin is also speaking at Mobilize) has discounted the PC completely, believing the future of social networking relies solely on mobile devices.

In handsets, device makers are grappling with new form factors and users interfaces as the phones original primary function, voice calls, falls to the wayside and the need to create a more immersive data experience comes to the forefront. New large-screened devices like Samsung’s Galaxy Note are blurring the distinction between smartphone and tablet, and my colleague Kevin Tofel believes that one day tablets will replace the smartphone entirely.

Nokia Siemens Networks’ conception of a heterogeneous network

On the network side, carriers and their infrastructure vendors have begun realizing that the big t0wer-based macro-umbrella networks that fueled two decades of voice services aren’t going to cut it in a data-centric world. They’re designing new types of small cells and base stations intended to deliver intense levels of bandwidth over limited areas. Those small cell deployments will eventually evolve into the new heterogeneous network, or HetNet, which will transform cellular systems from coverage-to capacity-focused topologies. Today’s carrier networks have tens of thousands of cells. Future networks will hundreds of thousands if not millions of cells.

The next few years are going to be tumultuous as we negotiate these seismic shifts from mobile voice to mobile data and from the PC-centric to the mobile-centric Internet. Not every MVNO, app developer and infrastructure maker is going to make it. We’ve already seen a big shakeup on the equipment side (Nokia decline and the dissolution of Motorola and Nortel Networks), and the big incumbent mobile operators are struggling to understand their role in the mobile broadband age.

At Mobilize, we’ll be tackling a raft of topics related to that transformation. Here are a just a few speakers to look out for:

• The technology gurus of two major carriers will be in attendance. Sprint CTO Stephen Bye and T-Mobile CTO Neville Ray will discuss their aggressive plans to overhaul their respective carrier’s networks. T-Mobile and Sprint aren’t just launching LTE, they’re actively shutting down large swathes of their 2G networks, and in Sprint’s case, pursuing new small cell and heterogeneous network (HetNet) architectures.
• Samsung has always been a mover and shaker in the handset market, but the rise of the smartphone turned it into the world’s largest mobile device maker. Samsung VP of Global Product Innovation Yoon Lee will discuss with GigaOM’s Stacey Higginbotham how build on that lead and challenge the ever-present Apple.
• Dealing with constantly fluctuating connection speeds – or no data connections – offline synching and huge array of different operating systems are problems with which the old internet didn’t have to cope. Parse CTO Kevin Lacker will discuss how mobile devices becoming the endpoints of the Internet has made the industry rethink the PC-centered models on which data services were built.

Featured photo courtesy of Flickr user Mark Strozier

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Mobile 2012 and beyond
• CES 2012: a recap and analysis
• What to watch in mobile in 2013