Push-to-talk: Kodiak Networks

One of the most remarkable things to note about push-to-talk technology is that at its core, it's a voice application that could become the hottest service of an era supposedly dominated by mobile data and content.

That irony is not lost on Craig Farrill, who was chief technology officer at an RBOC-affiliated mobile carrier during the industry's earlier voice era, and who is now a chief enabler of push-to-talk (P2T) for the carrier sector as founder and CEO of Kodiak Networks.

“This is the best thing that has happened to mobile voice in the last 12 years,” he said.

Farrill actually calls that “thing” instant voice rather than push-to-talk — not so much to avoid infringing on someone else's trademark, but to remind everyone how his own company's recipe for P2T, called Real-Time Exchange, is different from that of his competitors.

Kodiak favors a P2T architecture that leverages the inherent reliability of voice channels by combining voice-over-IP switching with transport of the circuit-switched voice networks. P2T solutions otherwise in use, including the Nextel Communications service that leads the market, use packet switching on data network channels. Nextel uses Motorola's iDEN scheme. Other solutions in use include the Motorola/Winphoria packet softswitch solution and fastmobile's server-based capability for smartphones.

“Voice over IP works. Voice over GPRS doesn't,” Farrill said. “VoIP was designed as wideband, GPRS was designed as narrowband.”

In taking advantage of voice network reliability, Kodiak's RTX also can give network operators more return on investment from these long-deployed networks.

RTX was first deployed by U.K.-based Orange and more recently had its U.S. debut with Alltel.

“[Alltel President and CEO] Scott Ford said Alltel has picked up 50,000 users in just two months, which is phenomenal, and triple what their expectations were for the service,” Farrill said.

Though the numbers are impressive, Alltel was not the first carrier to hit the market with P2T. Verizon Wireless and Sprint had the first P2T offerings to compete with long-dominant Nextel, but the late emergence of Kodiak's RTX as a technology option won't hurt the company's chance for carrier business if it consistently demonstrates that it's a grade above other technology options.

“Kodiak is not as late to market as one might imagine,” said Andrew Cole, senior vice president in charge of the wireless practice at Adventis. Cole said he has looked at the RTX on behalf of a carrier client that wants to deploy P2T.

“The push-to-talk stuff some carriers have deployed is not up to snuff, and they're still improving latency,” Cole said. “Kodiak could be a second-generation option that these carriers would be wise to look at.”

Though Farrill admitted Kodiak's technology currently can't beat Nextel on call set-up latency, it can beat all comers on intra-call latency by virtue of its use of full-duplex voice channels, he said.

“Verizon and Sprint have quit advertising P2T, and there's a reason for that,” Farrill said. “They have taken it back to the shop.”

Several carriers still have yet to decide on deploying P2T, though Cole said most major carriers eventually will feel the need to have it as a token offering. Most industry watchers think Cingular Wireless will be the next U.S. carrier to launch P2T. “Cingular desperately wants to get this service out at this point,” Cole said.

Bill Clift, chief technology officer at Cingular, recently said the carrier was evaluating its P2T options and was “looking at a technology from a company called Kodiak Networks.”

While Kodiak is the purveyor of the newest P2T technology in town, the company doesn't deserve full credit for P2T's status as a hot technology. Though Kodiak has a different and interesting take on the technology, there are other reasons to believe that P2T is this year's must-have technology.

For instance, many people in the industry believe it's a technology that can help increase network minutes of usage, as well as become an enabler of far more advanced services.

Messaging has become a wildly popular form of communication during the last few years, especially among younger demographics, and has contributed to increasing usage of data network facilities. However, P2T could have even more potential to prove the ongoing viability of voice communication.

“It's all about the amount of time you want to spend making a connection,” Farrill said. “If it takes you 27 seconds or more as a pretty fast typist to do it with messaging, but it takes you 7 seconds to do it with P2T, the choice is clear. We're trying to reduce the amount of time between thought and action.”

Even though it takes less time to use, the logic is that usage actually will increase. “It's been demonstrated that users of these services actually use their phones a lot more,” he said.

P2T architectures like Kodiak's that allow presence availability and management also can improve network efficiency and create a foundation for new services.

“Presence availability is the other big ‘a-ha!’” Farrill said. The ability to know if the person you are trying to reach has their phone on or off, or is too busy to handle the call, ultimately can reduce the number of missed calls, providing more effective use of network facilities.

Determining presence leads to the capability for other “instant” services. “Instant conferencing could be huge,” Farrill said. “You're talking about not having to call back and forth to sort through people's schedules to see if they're available.”

The ability to support that kind of evolutionary path is another reason why P2T could be considered a hot technology not just for this year, but for years to come.

“Presence availability is something that's nice to have right now,” Cole said. “It may not be a killer app yet, but it positions the technology for the future.”

After being adopted by Nextel's more direct competitors, Farrill sees more small carriers in rural markets wanting to adopt P2T technology. He thinks Kodiak's voice network-centric solution will prove an attractive option because some of these carriers have not progressed very far yet in their data network upgrades.

He also said that more international carriers will begin to adopt the technology to serve the needs of multinational enterprise customers, making P2T one of the few wireless technology evolutions of recent years that will spread from west to east, rather than the reverse.

Cole said Farrill's status as a former carrier technology chief gives Kodiak a leg up in selling its solution to these service providers.

“Without that past carrier expertise, it can sometimes be difficult to sell to the carriers. It can go more slowly because there might not be as much direct understanding.”

As the market for push-to-talk surges behind deployment from these later adopters, Cole also said the vendor competition for carrier deals probably will remain tight among a few vendors, unless another company comes up with a startling new technology innovation on the P2T front.

“There's not room for other vendors unless it's a quantum leap change from what's out there now,” he said. “It could only happen if someone developed a technology platform that could do a number of things, but which also enabled push-to-talk.--Dan O'Shea

WiMAX: Intel

There are a lot of eyes on Scott Richardson — not only eyes from his own company, but also eyes from vendors and carriers the world over. They're eagerly anticipating what the broadband wireless unit of Intel, of which Richardson is general manager, will do with WiMAX.

The initial expectation is for Intel to produce the next generation of its Centrino mobile chipset for WiMAX. But the great hope is that Intel will accomplish what no other vendor in the broadband wireless space has done: to create a chipset that will pervade mobile computing and, in essence, build an industry around WiMAX.

Those may seem like high expectations for a lone company, but there is precedent. Intel gave an enormous boost to Wi-Fi technology when it launched Centrino more than a year ago. Its low-power integrated 802.11 chipsets made Wi-Fi almost ubiquitous in higher-end laptops, and helped foster the belief that wireless connectivity is almost a given in portable computers — just like a CD-ROM drive.

Now it's up to Richardson to repeat that feat with WiMAX, and he has little time to waste. Intel has pledged to have its first Centrino global WiMAX chipset out by the end of 2006. “It's an aggressive goal, but we're working hard to finalize the standard,” Richardson said.

After launching Centrino last year, Intel immediately began to look to the future to where the technology would evolve. Richardson's team quickly realized that Wi-Fi could only go so far in supporting the wireless ISP and carrier hot spot business models that the industry began producing. Another evolution in wireless technology was needed to create a true broadband wireless experience, Richardson said.

“WiMAX is really what we consider carrier Wi-Fi,” Richardson said. “If you asked carriers what they would change about Wi-Fi, they would basically describe WiMAX.”

Unlike Wi-Fi, however, Intel dove into WiMAX early. When Intel launched Centrino, Wi-Fi had been fully standardized for years, and certified equipment was available for some time (Intel couldn't make up its mind between Wi-Fi and competing standard Home RF). But with WiMAX, Intel chose early — selecting the IEEE 802.16e platform over other technologies such as OFDM's 802.20 — long before the WiMAX standard was finalized and equipment certified.

WiMAX was recently finalized, leaving the first half of 2005 for certification trials and the last half for commercial rollout. Given that timeline, Intel will have its first chipset out a year after WiMAX's commercial “birth” — an accelerated timeline, considering that WiMAX base stations need to reach certain penetration levels before computer makers can justify installing WiMAX chipsets into every laptop.

But when that day comes, expect Intel to make a massive investment into the technology, said Patrick Zerbib, vice president of Adventis.

Intel accepted significantly reduced margins — around 50% — on the Wi-Fi portion of its Centrino chipset to drive sales of its processors, Zerib said. Its logic was that if it can create a new market for wireless, the company could create a whole new market for its high-end mobile processors where margins are in the 80% range, he said.

“It's a very smart strategy,” Zerbib said. “Intel recognizes that the entire digital arena is changing. Creating that market for WiMAX won't be an easy slam dunk for Intel, but this company is so powerful and they have so much cash — if anyone can push this technology, Intel can.”

Intel's commitment to wireless is undeniable. While Intel has released no figures on how many Centrino chipsets it has shipped, the company estimates it has spent more than $400 million marketing the technology and has made $4 billion in capital investments in wireless technology in general. Intel seems primed to commit those same resources to WiMAX, but even with that degree of guidance it's difficult to predict what path the technology will follow — whether it will be a technology for WISPs using unlicensed spectrum in mature markets, or licensed operators looking for a DSL substitute.

While Intel is counting on making its claim in the 802.16e portion of the WiMAX specification, the section delineating guidelines for portable low-power devices, that could easily change, Richardson said. Even with rollout of Wi-Fi — after the market had several years to weigh the technology — Intel was surprised at all the twists and turns the technology made. Intel had imagined Wi-Fi to be more of a home networking and business LAN technology. It became those things and much, much more, Richardson said.

“We have never underestimated the fact that laptop users want to have wireless access,” Richardson said. “What surprised us were all the new business models that wireless access created. We never imagined all these wireless hot spot businesses would pop up. In many ways, our whole WiMAX strategy is designed to feed the demand that Wi-Fi created.--Kevin Fitchard

EV-DO: Qualcomm

With all apologies to Monet Networks and its groundbreaking but ultimately failing plan, Verizon's announcement that it would deploy EV-DO technology in portions of its network this year generally might be considered a defining moment for the technology.

Or maybe it was two Korean carriers launching service last year. Neither really qualifies, though, according to Jeff Belk. Having worked as something of an EV-DO evangelist for the past several years, the senior vice president of marketing for Qualcomm has been pushing carriers to see the benefits of deploying the high-speed technology and why it offers more than EV-DV, WCDMA and now Wi-Fi.

“There's a very consistent process to the way technologies get developed regardless of the spin cycle,” Belk said. “What DO is about — and frankly what UMTS is about — is fundamentally altering the consumer experience.”

Indeed, what makes EV-DO a hot technology isn't necessarily what it does in a lab environment, but how it is performing in the real world. (On paper it doesn't look too bad, with a theoretical top speed of 2.4 Mb/s, and real-world experience putting it closer to 300 kb/s.) As originator of EV-DO, Qualcomm is providing the underlying technology to virtually every current deployment. Lucent Technologies, which is the main supplier of Verizon's EV-DO equipment, is using Qualcomm chip sets in several portions of the network.

Of all the high-speed mobile data technologies currently on the market, EV-DO is among the most widely deployed, with Korean carriers KT Freetel and SK Telecom, Brazil's Vesper, KDDI in Japan and Taiwan's APBW all in deployment mode.

Perhaps more telling is that in several of its multiple filings related to the acquisition of AT&T Wireless, Cingular acknowledged that EV-DO has no peer in the current market and that to compete, the carrier must look at the same technology. “U.S. consumers are now demanding capabilities that require large amounts of bandwidth at high speeds to work properly. Cingular and AWS currently use data transmission technologies such as GPRS and EDGE that are unable to accommodate all these demands,” one filing reads.

Ultimately, the success of any data technology is measured by user adoption and the ability to provide carriers with new revenue-generating services. On both counts, the early numbers for EV-DO are promising.

“You can look pretty much everywhere that DO has been deployed and see they're getting strong ARPU boosts,” Belk said.

In Korea, for instance, SK Telecom reported more than 1 million EV-DO users within eight months of its launch, while KTF claimed around 700,000 in roughly the same period. On an ARPU basis, the numbers are better still with SK Telecom reporting $14 from EV-DO services and KTF reporting $10, according to a recent research note from Ovum's Asia Pacific research director Rosalie Nelson.

Part of the reason the technology has taken off is its ability to give users their first experience in the mobile environment that approximates the broadband experience in the home. Among the most popular applications in the first EV-DO rollouts are sending video clips and watching short video news stories, according to Nelson. Without the raw speed of EV-DO, that simply wouldn't be possible.

“CDMA2000 and EDGE can start to do multimedia and show the way,” Belk said. “But from a consumer experience, they say, ‘This isn't as fast as my broadband at home.’ DO is for the first time giving people something close to that experience.”

By coincidence of timing, though, EV-DO proponents are spending as much time these days defending the technology against an incursion of mindshare from Wi-Fi as they are hyping the benefits of the protocol. In some cases, Wi-Fi advocates are claiming that their technology — and its cousin, WiMAX — will be able to function as less expensive technology alternatives to EV-DO.

But Wi-Fi's lack of reach in real-world environments and lack of business model just can't compare, Belk said. After three years of hype, the number of actual public coverage area of Wi-Fi is relatively small despite the number of hot spots.

“From an aggregate coverage perspective, it's about the size of one suburban cell site,” he said. “Consumers want to be connected as fast as they can, everywhere, for the least amount of money. Wi-Fi works under this assumption that people are comfortable driving 5 to 10 minutes, and that's not necessarily true.”

Belk is backed up by a study from Alexander Resources that claims Wi-Fi is more of a complement to EV-DO, better suited to service notebook PC users in select airport, hotel and restaurant locations. The addition of voice over Wi-Fi doesn't change the business case either, said Jerry Kaufman, president of Alexander Resources.

“When technologies like EV-DO get deployed, a lot of people are going to understand that the Wi-Fi business model is going to get a lot worse,” he said. “There's just no way anyone is going to be able to compete on price and services with technologies like EV-DO. It's really a customer interface issue.”

Such analysis is music to Belk's ears. And in fact, much of the promotion campaign around EV-DO focuses on user experience more likely to appeal to mass-market consumers and less on raw speeds that thrill the techie early-adopter crowd.

Case in point: As the cameras in phones edge up into the 3- and 4-megapixel area, the ability to transfer images over DO networks will give it an advantage over other technologies and transform the way users view phones.

“What DO and other really high-speed technologies allow people to do is turn what were products like digital cameras and MP3 players into features,” Belk said. “And once you give people high speed, you can't go back.--Vince Vittore

Content Platforms: Sun Microsystems

Sun Microsytems' David Rivas may not be a mobile content junkie, but as the chief technology officer of the company's consumer and mobile systems group, he “is pretty committed to this whole digital lifestyle thing.”

Rivas is responsible for the overall technology direction of the group. It's more of a “where technology meets strategy” role, he said. For him, the mission is simple: Bring mobile content to the consumer.

He said his double-E helps. So does his foray into research on models for computer music, his experience as the architect for the Java Media Framework and his role driving standards for Java interactive television. “All this stuff touches the consumer, and that's the kind of environment we're focused on now,” Rivas said. “Handsets in particular are primarily consumer devices these days, so the Java experience is quite relevant to that.”

The Java experience, from Sun's point of view, has been pretty good so far when it comes to mobile content. Java-equipped handset sales tripled in 2003 to 95.5 million. It wasn't a bad experience for the industry, either. Java contributed $1.4 billion in revenue to operators worldwide, according to a recent study by the ARC Group, which predicted that could grow to $15.5 billion by 2008.

However, it takes more than devices to deliver content, and the mobile content distribution market is still up for grabs. As a matter of fact, it's still not all that clear who the players are in the mobile content distribution market.

In addition to the operators and content providers on either end of the delivery model, content aggregators and a myriad of infrastructure providers are beginning to jam the middle.

“And there's no clear distinction between the categories,” said Adam Zawel, analyst for The Yankee Group. Aggregators will sometimes sell infrastructure directly to an operator, while companies like Sun often reach out to content providers and aggregate content for them, he added.

The good news is that operators will be buying. Zawel said 10% to 40% of operators' revenue would soon come from data, depending on their geography. “A substantial chunk of operator investment in support of data will be in the core network, but a lot of investment will be for managing and delivering the content.”

Sun will be trying to capture that revenue from the device end to the infrastructure end as well as the partner management processes in between. But because roles have not been clearly defined and operators have not yet determined their preferred method of content distribution, partnership will rule the day.

Sun is nothing if not one of the most prolific partners in the industry. So it was no surprise when the company stepped in to manage a burgeoning cadre of device manufacturers and content providers that were organizing around the challenge of content certification and developed what is now the Java Verified Program.

Sun, Motorola, Nokia, Siemens and Sony Ericsson launched the Java Verified Program in February. Its goal is to provide application developers a more direct route to market. It offers a set of guidelines and tests an application must meet in order to be included in the companies' product catalogs. Wireless operators can then access these applications for distribution to its subscribers with confidence it is a compliant product.

“You can think of it as a funnel where on the top end we are bringing in the developers and making their content available to operators and other catalogs,” Rivas said.

While participating companies have consolidated their test and verification efforts in the name of cost and efficiency, many still manage their own content catalogues — including Sun, which offers content on its Java.com Web site.

“It has become apparent to carriers and content creators that they can't continue to take their new applications to all the carriers and phone manufacturers to get approved,” said a Nokia spokesman. “So this industrywide testing platform is a great streamlining initiative.”

Nokia, which had its own verification platform, has migrated its processes over to the Java Verified Program. It still operates its Nokia Trade Point content catalogue.

Testing and verification streamline only one part of the process. For the distribution component, Sun has integrated several server and software products into the Java Enterprise System (JES).

“In order to actually deploy a service, one needs to be able to construct a service and to construct a service you need infrastructure software,” Rivas said.

The systems that make up the JES include application servers, directory servers, communication servers, identity servers and authentication servers. Sun sells them separately but has fully integrated them, so that when a new software release comes out, it comes out for all. Each server also is interchangeable with other vendors' Java-based servers.

Sun also has instituted a per-head pricing plan that is based not on the number of subscribers or users, but on a company's employee head count.

“It provides tremendous leverage for companies providing services to a large customer base. They don't have to swallow a gigantic software bill,” Rivas said.

A second piece of the infrastructure is Sun's Content Delivery Server. The new product, which will be the cornerstone of the company's content distribution efforts, originated with Sun's acquisition of Pixo last June.

San Jose, Calif.-based Pixo developed Java-based server software that managed the secure distribution and monetization of digital content for mobile devices. It is the piece of the puzzle operators need to catalogue their content, manage the fulfillment and purchase of the content and integrate with the billing system for authentication and purchasing rights, Rivas said.

Even with the right infrastructure in place, devices can act both as the enabler of mobile content and an inhibitor to adoption. Memory, power and display size all contribute to limiting the type of content that can be offered. And there is the competition. Qualcomm's Binary Runtime Environment for Wireless (BREW) shipped 11.6 million devices last year, up from 3.5 million the year before, and is expected to ship 75 million by 2008. The technology practically owns the Asia-Pacific market.

The Yankee Group's Zawel said Qualcomm has done the best job so far of wrapping up the entire content experience and delivering it in a box to carriers.

“It's everything you need between the content and the carrier — it's just not open enough,” Zawel said. On the other hand, he said, “Java is a more open environment but is also is more chaotic.”

Out of the chaos of thousands of content developers and a broad range of partners, however, may come the serious application that will supplant gaming as the leading choice of content applications.

One such application recently released on a Java platform is the Pulse Meter mobile health solution from MedicTouch. The application lets users monitor their pulse, view the results on a Java-enabled high-resolution device and transmit the data to a Java-compliant server.

It is this sort of end-to-end scenario where Java enables every touch point that Sun is counting on.--Tim McElligott

4G: Flarion

People who work in telecom are accustomed to being skeptical of technology trials. After all, carriers' labs are often bigger than their wallets, and even a field trial is no firm guarantee of commitment.

But when Nextel Communications expanded its trial of Flarion's Flash-OFDM broadband wireless system in mid-April and launched it commercially in the Raleigh, N.C., area, it ignited new buzz around the six-year-old company.

Nextel effectively doubled the original size of its Carolina trial to include 127 cell sites, where it covered about 1300 square miles and about 1 million people. After two months of testing with 800 users, Nextel found Flarion's technology could deliver 1.5 Mb/s downstream (with burst rates of up to 3 Mb/s) and 375 kb/s upstream (with 750 kb/s burst rates) — comparable to DSL and cable modem, but with a mobility offered by neither.

“What is magical about this — which doesn't happen too often in our industry — is the brochure claims stack up to the performance,” said Flarion President Mike Gallagher. “Once you've gone to a commercial network, anyone can experience it. The question of, ‘Does it work as advertised?,’ which is always a tough one, rings true.”

Analysts seem to agree. Some took the launch of the commercial service as a clear sign that Nextel will deploy Flarion on a wider scale.

“We are assuming Nextel will move forward with a 3G/Flarion buildout and that the company will spend $2 billion to $3 billion in capex/operating expenditures over a three-year period,” Thomas Weisel Partners wrote in a report published one week after the Raleigh trial expansion. “Such a move would put the company in [a] very good position to compete with Verizon Wireless' 1x EV-DO network and be ahead of the game potentially vs. [Sprint] PCS and T-Mobile, as well as the combined/distracted AT&T Wireless/Cingular.”

However, Nextel continued to test rival technologies all the while. “Don't let Nextel fool you — they're sleeping with everybody right now,” said Ed Rerisi, vice president of research at ABI Research.

In fact, Nextel wasn't trying to fool anyone. In the company's first-quarter earnings conference call that month, CEO Tim Donahue said plainly, “Nextel continues to evaluate other broadband technologies, particularly as they apply to the 2.5 GHz band. I want to be perfectly clear when I say that no broadband decisions have been made to build out the next-generation network at this time.”

Still, the Nextel trial gives Flarion the distinction of supporting the only major U.S. carrier with a commercial service of this kind. One rival, Navini Networks, claims customers in nearly two dozen countries. But in the U.S. — although it has been tested by BellSouth and Sprint — most of Navini's customers are wireless ISPs. Another competitor, IPWireless, has been in trials with NextWave Telecom and Texas WISP Clearwire Technologies, and its gear provides broadband in several Maui, Hawaii, hotels — but nothing like Raleigh. At press time, IPWireless was expecting to enter a field trial with a major U.S. operator soon, said marketing director Jon Hambidge, adding, “We're going after carriers, not 2000 wireless ISPs.”

Flarion representatives seem to loathe comparison to the other two with which it's frequently named, dismissing them as fixed wireless players without the sophistication to master true mobility.

For example, when asked to respond to an analyst's charge that Flarion and IPWireless were too similar to clearly differentiate, Gallagher replied, “That's not worth even commenting on.” Similarly, IPWireless' Hambidge scoffed at Flarion's claims of superior mobility, claiming that his company's 100 millisecond handoffs between cell sites are “the best in the business” and that each IPWireless base station provides four to five times the capacity of a Flarion base station.

Navini caused a stir recently by reversing previous public statements and joining the WiMAX Forum, a group that supports the IEEE standard 802.16 for broadband wireless. But according to a recent report from the Gartner Group, the mobile version of WiMAX — a technology that is going through a complex standardization process consisting of several stages, of which mobility is one of the latest — won't be used widely until 2009 at the earliest.

“I don't think people are going to wait around to see mobile WiMAX working,” said Hambidge, arguing that IPWireless' allegiance to the 3G Project Partnership (3GPP) UMTS TDD standard gives it a better shot at intercarrier roaming than Flarion's technology.

“We have no intentions of keeping this a proprietary technology,” Flarion President Mike Gallagher said, without mentioning the nascent 802.20 standard Flarion supports and that some believe may one day be the standard for Flarion's technology. Development of 802.20 has stalled amid disagreements among stakeholders, making the proprietary approach seem to some like the best bet at the moment.

“There's lot of dissension right now in 802.20,” Rerisi said. “Some of these companies may try to go at it on their own instead of trying to apply for a formal standard.”

According to ABI's Rerisi, Flarion's biggest weakness at the moment is its lack of an OEM partnership with a large incumbent equipment vendor. That wound was salted in April when Navini announced an OEM deal with one of its investors, Alcatel (which already has been reselling IPWireless gear). Combined with that news, the IPWireless partnerships with UTStarcom and Samsung leave Flarion looking like the pretty girl who doesn't have a date for the prom.

But Flarion execs say the best way to get an OEM partner is to win over carriers first, which will motivate those carriers to, in turn, pressure their incumbent vendors to work with Flarion. And Flarion already has “partners” of a sort: Nortel installed and serviced the Nextel deployment, and Motorola assisted Flarion with its showcase network for safety personnel in the District of Columbia, which should go live in coming months. For OEM partners, just stay tuned, Flarion said.

In the meantime, anyone with an interest in mobile broadband will no doubt keep an eye on Nextel and North Carolina. But although that deployment has made Flarion highly fashionable this year, Gallagher attested that its Flash-OFDM is no flash in the pan.

“The old name of the company was Radio Router,” Gallagher said. “That's really what we are: a router for radio. Mobile broadband is exactly what this company was built for six years ago.--Ed Gubbins