Illustrations of Network Virtualization from VMware, ETSI NFV, and Intel Reference Designs


Our last article described VMware’s view of network virtualization, provided definitions and advantages of same, with different opinions expressed on the main advantage(s).  There was a reference to a follow on article which would illustrate various configurations of network virtualization.  We attempt to do that in this article by describing figures from VMware, ETSI NFV ISG, and Intel’s SDN/NFV reference designs.

Network Virtualization Illustrations

1. VMware

The figure below (from VMware) shows a Leaf/Spine L3 fabric deployment in support of Network Virtualization. The fabric connects compute cabinets with Hypervisors (below), Infrastructure cabinets with controller nodes, and Edge cabinets that interface to the outside world (e.g. the Internet, private lines, IP VPNs, Carrier Ethernet, etc).

An image is depicted showing leaf-spine architecture for network virtualization.
Image Courtesy of VMWare.

Because the virtual network is decoupled from the Data Center switch fabric, the latter can be built without the former  complicating or restricting its design.  VMware believes that the most scalable, robust, and cost effective architecture (to date) for such a Data Center switch fabric is the Layer 3 (L3 or IP Network layer) Leaf/Spine fabric design shown above.

Such a L3 Leaf/Spine fabric is constructed using standard IP routing protocols (e.g. OSPF, IS-IS, BGP) between the Leaf and Spine switches. This fabric can be put together using commonly available IP networking equipment such as L3 switches that support IP forwarding and 1/10/40G Ethernet MAC framing.

A Leaf switch is connected to all Spine switches to provide multiple high bandwidth paths to any other rack. The Leaf switch selects a path for each new flow between any pair of Virtual Machines (VMs). That’s done in hardware at line rate (e.g. 1/10/40G b/sec). This path selection is referred to as Equal Cost Multi Path (ECMP), which is supported by any standard and commonly available L3 switch. The selected Spine switch receives the traffic from the Leaf and forwards to the destination Leaf based on IP routing (looking at the destination IP address in the tunnel headers).

The Hypervisor nodes, running a programmable vswitch, attach the Leaf switch like any standard server — with a Network Interface Card (NIC) that has an IP address.  The IP address on the NIC is used to dynamically build tunnels between other Hypervisors and Gateway nodes. The NSX Controller programs these tunnels dynamically as the environment changes.

Note that there is no special protocol between the Hypervisor and Leaf switch; just IP/Ethernet frames. If there was NIC Ethernet PHY bonding used on the Hypervisors, then Ethernet Link Aggregation Control Protocol (LACP) would be required.

2. ETSI Network Function Virtualization (NFV) Industry Specification Group (ISG)

The ETSI NFV ISG’s charter is to issue recommendations that will be input into existing Standards Development Organizations (SDOs) like ITU-T, along with industry forum like the ONF.

The figure below illustrates the NFV reference architecture. This diagram serves as a starting point for the NFV Architecture Working Group, but has not yet been finalized. NFV is broken into broad functional domains including the Applications Domain (where Network Functions reside), and the underlying framework, consisting of the HyperVisor, Compute, Infrastructure Network, and Management and Orchestration domains. The NFV architecture explicitly is defined to be complementary to SDN. However, recognizing the early stage in the SDN life-cycle,  it is desirable to realize the benefits of NFV based on existing network architectures.

Image Courtesy of Open Networking Foundation

3. Intel Reference Designs

Intel recently introduced three referenced designs targeted at SDN/NFV implementations for both the control and data planes.

The Intel® Open Network Platform Switch Reference Design
Codenamed “Seacliff Trail,” the Intel® Open Network Platform (ONP) Switch Reference Design is based on scalable Intel processors, Intel® Ethernet Switch 6700 series and Intel® Communications Chipset 89xx series, and is available now. The ONP Switch Reference Design will include Wind River Open Network Software (ONS), an open and fully customizable network switching software stack using Wind River Linux. Wind River ONS allows for key networking capabilities such as advanced tunneling as well as modular, open control plane and management interface supporting SDN standards such as OpenFlow and Open vSwitch. Common, open programming interfaces allow for automated network management, and coordination between the server switching elements and network switches enabling more cost-effective, secure, efficient and extensible services.

The Intel® Data Plane Development Kit (Intel® DPDK) Accelerated Open vSwitch
Network architectures have traditionally been optimized for large packet throughput to meet the needs of enterprise end-point applications. Intel is executing a project aimed at improving small packet throughput and workload performance that can be achieved on the Open vSwitch using the Intel DPDK. Intel is specifically re-creating the kernel forwarding module (data plane) to take advantage of the Intel® DPDK library. The Intel® DPDK Accelerated Open vSwitch is planned to initially be released with the Intel® ONP Server Reference Design in the third quarter of this year.

The Intel® Open Network Platform Server Reference Design

Image courtesy of Intel
Image courtesy of Intel

This server reference platform, codenamed “Sunrise Trail,” is based on the Intel® Xeon® processor, Intel 82599 Ethernet Controller and Intel Communications Chipset 89xx series. The ONP Server Reference Design enables virtual appliance workloads on standard Intel architecture servers using SDN and NFV open standards for datacenter and telecom. Wind River Open Network Software includes an Intel DPDK Accelerated Open vSwitch, fast packet acceleration and deep packet inspection capabilities, as well as support for open SDN standards such as OpenFlow, Open vSwitch and OpenStack. The project is in development now: the first alpha series is slated to be available in the second half of 2013.

Intel does NOT distinguish between SDN and NFV in their whitepaper, which describes how to use their three reference designs.  Intel’s  SDN/NFV architecture consists of four layers: orchestration, network applications, network controller, and node-as shown in the figure to the right.

These layers have been proposed by Intel and have not yet been accepted (or even submitted as a contribution) by ETSI NFV ISG or the ONF which is standardizing SDN-Open Flow. Further details are in the Intel whitepaper referenced above.


It should be quite obvious to users that VMware’s network virtualization is an actual implementation, while the ETSI NFV and Intel illustrations of network virtualization are high level architectural diagrams, i.e. at the concept stage.  That’s what one would expect in this very early process of standardizing network virtualization, with no solid specifications likely for at least one or two years.

Please read the comments underneath my last article for different opinions, perspectives and links to relevant ETSI NFV specification work.

Intel to battle Qualcomm for mobile computing leadership

Update from Intel Development Forum in S.F. and Qualcomm open house in Santa Clara, CA last week.


For those interested in mobile computing and mobile WiMAX, there were several important take-aways from last week’s Intel Developer Forum (IDF) in San Francisco. CA. Most importantly, Intel made it clear that they intend to dominate the mobile computing silicon and software space. But Intel will face stiff competition from Qualcomm, which is already moving their technology into smart phones, eBook readers, Mobile Internet Devices, and other hand held devices.

In this article we detail Intel’s newly announced mobile computing plans and its reaffirmation of its commitment to mobile WiMAX. We’ll also look at Qualcomm’s plan for mobile computing dominance. We will examine the following:

  • Roadmap for Intel’s Atom processor, which is targeted at netbooks, MIDs, and other hand held mobile computing devices.
  • Establishment of the “Intel Atom Developer Program.”
  • A new version of Moblin – Intel’s Linux-based operating system intended for netbooks and MIDs.
  • Reaffirmation of Intel’s WiMax commitment via a short progress report and Clearwire demonstration.
  • Qualcomm’s mobile computing intitiatives

Editors Note: More information on this year’s Intel Developer Forum can be found at:

IDF Highlights:

1. Intel’s Atom processor (now using the Menlow platform) is currently sold into netbooks, but not hand held devices where power consumption is a huge issue. Intel plans to integrate more functionality into Atom, reduce power consumption, and shrink the die sizes by half.

Editors Note: Atom and other “System- on a -Chip (SoC)” platforms have their own road map schedule that is different from Intel’s PC client and server products, which operate on Intel’s famous "tick-tock " road map schema. In particular, the Atom/SoC line will trail the leading PC and server products by about one semiconductor process generation, or "tick." Major architecture upgrades — or "tocks" — also will be different from the PC client and server road map.

The new Moorestown platform –to be available in mid 2010- shrinks and integrates more functions into fewer chips so that it can be embedded into Mobile Internet devices (MIDs) and larger smart phones. It also adds hyper-threading to Atom, boosting performance in threaded apps. Built on a 45nm process, Moorestown will greatly reduce power consumption (especially idle power) compared to Menlow. Citing a combination of architectural, design and process enhancements, Intel VP David Perlmutter discussed some of the innovative techniques that Intel is implementing in Moorestown. These include “Distributed Power Gating,” for improved performance and major reductions in power and thermal envelope.

Mr. Perlmutter also highlighted Intel’s next generation of mobile processors, codenamed "Arrandale," which brings the Nehalem micro-architecture to mainstream laptops. These chips will integrate the dual-core CPU and graphics in the package and incorporate the 32nm manufacturing process and second-generation high-k metal gate transistors for increased performance and power efficiency for mainstream mobile PCs. This integration of platform components will continue into the future with a fully monolithic processor on 32nm, codenamed "Sandy Bridge."

These technologies help to achieve up to a 50x improvement in platform idle power reduction compared to Intel’s first generation "Menlow" platform. The reductions are enabling Intel to establish new thresholds in ultra low power while making it possible to run the full Internet and media-rich applications in handheld devices, according to Perlmutter.

In his opening day keynote speech, CEO Paul Otellini explained, “At Intel, Moore’s Law is alive and thriving. We’ve begun production of the world’s first 32nm microprocessor, which is also the first high-performance processor to integrate graphics with the CPU. At the same time, we’re already moving ahead with development of our 22nm manufacturing technology and have built working chips that will pave the way for production of still more powerful and more capable processors."

Editors Note: It wasn’t too long ago that 90nm processors were the norm, and 65nm processors were cutting edge.

For sure, Intel is serious about pushing its Atom-based SoC platforms into an increasing number of product categories. CEO Otellini even predicted "a future where Intel ships more SoC cores than standard PC cores."

2. The Intel Atom Developer Program was unveiled during a keynote speech by Renee James, corporate vice president and general manager of Intel’s Software and Services Group. This effort encourages independent software vendors and developers to create mobile applications. Intel is partnering with notebook and netbook manufacturers, including Acer and ASUS, to create multiple application stores where applications and application building blocks for Intel-based netbooks and handhelds will be sold.

"The Intel Atom Developer Program will drive new innovative applications for consumers and new revenue streams for software developers," said Mr. James. "The new program facilitates the creation of applications that consumers can use with confidence knowing they were optimized and validated for Intel Atom processor-based devices. We want to fuel the growth of Intel Atom-based products designed for the mobile lifestyle," Continuing, he stated, "The netbook has become one of the most popular consumer devices in the market today, but its true potential has been limited by applications that are not optimized for its mobility and smaller screen size. The Intel Atom Developer Program provides a great opportunity for developers to create useful and inventive applications that will unlock a netbook’s potential while opening a new sales and distribution channel."

To provide the broadest choice of applications across platforms, the Intel Atom Developer Program will support multiple operating systems and run-time environments. Run-times enable developers to use a single code base to support multiple device platforms and avoid extensive reprogramming, thereby reducing development costs and time-to-market. Run-times such as Microsoft Silverlight allow developers to access multiple classes of customers and deliver rich applications for Windows and Moblin-based environments using a single toolset, Visual Studio and the .NET Framework.

For more information, please see:

Intel no doubt wishes to replicate the success of the iPhone App Store by having its OEM vendors create App stores for their Atom-powered devices. Taking it one step further, Intel introduced its own "App Store" for its Linux-based Moblin operating system (see discussion in 3. below). Interested developers can join the program now at this new site:

Intel states, “You create innovative network applications – we take care of the rest.” Let’s see if this works out well for all concerned.

3. Intel introduced a new version of Moblin – its Linux-based operating system for netbooks, MIDs and other hand held devices. Among several new features, Moblin 2.1 supports a new interface appropriate for the screens you see on MIDs and smart phones.

Additionally, Moblin 2.1 builds in capabilities like native touch screen input and gesture support, new user interface features, and support for more hardware drivers. It also includes incremental upgrades that expand the usability of the OS.

Moblin was originally developed and promoted by Intel as an OS for netbooks. At IDF, an Intel representative said that with Moblin 2.1, the OS will now come in three versions: for handhelds, netbooks and nettops. (A nettop is an inexpensive desktop about the size of a hardcover book to which key boards, mice and monitors can be connected.) For more information, please see:

4. Intel is still very committed to mobile WiMax—even as global mobile carriers rush to implement various versions of 3G and look towards LTE in the future. During his keynote speech, David Perlmutter, executive vice president of Intel’s mobility group, stated:

“Another cool factor is connectivity. This is the era of getting connected. Intel is delivering the total mobile experience on each device, offering different levels of performance and power in sleek form factors coupled with compatibility, a superior mobile Internet experience and embedded WiMAX wireless broadband. We’re truly taking mobility to the next level of cool."

Perlmutter continued,

“A laptop generates more than 15 times the data of a smart phone. This is generating stress on 3G networks. Last year we introduced the first embedded WiMax that goes into notebooks. But having products with WiMax doesn’t matter. It’s all about having networks, because if you cannot connect, then it doesn’t matter. And we have networks being built in North America, in Russia, in Japan, and we have networks already there. We are building with our partners networks in other places like India, Malaysia, Taiwan. And many, many other places have all sorts of mobile and fixed WiMax all over the globe.”

In closing this discussion topic, Perlmutter said, “But I will better have someone that build networks talk about it.” Peter Cannistra1, VP of embedded devices at Clearwire was called to the stage. Mr. Cannistra talked about his company’s “ability to put the mobile Internet in your hand, in a broadband way.” In particular, Clearwire customers are realizing average downstream speeds of 3 to 6 M b/sec, bursting up to 10M b/sec, according to Cannistra. He then demonstrated a videoconference over WiMAX, which showed superior performance compared to an equivalent 3G videoconference. The comparison was a CLEAR win for mobile WiMAX (pun intended).

Perlmutter’s keynote webcast includes this videoconferencing comparison as well as his comments about the Atom processor road map. It is archived at:

Qualcomm’s Mobile Computing Initiatives

As we all know, notebook PCs are evolving into netbooks, tablet PCs and MIDs, essentially big smart phones that are always on, always connected to the Internet, with all-day battery life — in short, very much like a large iPhone or BlackBerry. Qualcomm calls these devices smartbooks because the design resembles a large smart phone.

The cellphone’s guts have been the domain of Qualcomm. As the cellphone becomes more like a computer and the computer more like a cellphone, it was inevitable that Intel and Qualcomm would battle for dominance of mobile computing.

“Intel is trying to come down from the computer and bring their software ecosystem along,” said Qualcomm’s chief executive, Paul E. Jacobs. “We’re trying to go up from the phone and build the software ecosystem.” Mr. Jacobs sees his company at the center of an industry that is driving the most cutting-edge innovations, as seen in devices like the iPhone and BlackBerry Storm.

“That energy is now coming out of the phone industry,” Mr. Jacobs said. “The PC became so standardized that the degree of innovation was not the same as what you see in the phone space.”

In an open house discussion on September 24th, Qualcomm executives described the many software initiatives underway.  The mobile web is a key focus for Qualcomm.  In a follow up email exchange, VP of Research Nayeem Islam wrote:

"Qualcomm has many initiatives in wireless. Clearly, we have fundamental research in basic communications but that is in San Diego. As you can tell from the talk yesterday we have a very large initiative in wireless and a very strong focus on mobile computing based on our snapdragon platform. We have also started a rather large fellowship competition in new areas like web technologies (the focus of last night’s talk),  indoor positioning technologies, visual computing and augmented reality, multicore and low power computing, gaming, application delivery, and content delivery. You can learn more about our fellowship at: "

For another point of view, please see:

Comment and Analysis:

There is no doubt that Intel is focusing on mobile computing with its next generation of microprocessors, Atom developer program and new version of Moblin. Will they be able to compete with future versions of ARM or Qualcomm’s Snapdragon with embedded 3G cores?

Until now, the key issue has been power consumption, with Intel’s Menlow platform consuming too much power to be used in handheld devices. Intel’s only hand held design win that we are aware of is with LG – a 3G MID that will use the Moorestown platform. No deliverable products have been announced for the Intel- Nokia strategic partnership. We expect Moorestown to be used with embedded 3G SoC capability with 3G cores from Ericsson, Nokia and possibly other vendors.

So we think the key test, will be commercial acceptance of Intel’s "Moorestown" platform, scheduled for 2010, in eBook readers, MIDs and smart phones.

With Qualcomm now making a big mobile computing push with its Snapdragon (ARM based) processor and with their Gobi 2000 platform that supports almost all 3G wireless network interfaces we think they will be a formidable competitor. At an open house last week, we learned that Qualcomm is producing a wealth of software innovations to advance the always-connected mobile world. The mobile phone chip powerhouse now has more software engineers than hardware engineers and is moving ahead full steam with mobile computing research projects. Qualcomm is mobile OS neutral, but is currently working with many device vendors using the Android platform.

A key differentiator between the two companies is that Qualcomm is mobile OS neutral, while Intel is trying to promote its Linux based Moblin OS. This is somewhat surprising as Intel’s “strategic partners” (e.g. Nokia) use a different Operating Systems, e.g. Nokia’s Symbian, Android, Windows Mobile, Palm OS, etc. This is quite perplexing panoply of mobile OS’s that app developers have to contend with.

And with the potential for Microsoft to port Windows 7 to ARM-based platforms in the future, we believe Intel could potentially devote even more resources to Moblin in order to help deter Microsoft from supporting Windows on ARM as well as Android on Snapdragon. The Atom Developers program will also need to be strengthened in order to offer a variety of applications that will run on the Moblin platform. Can this initiative succeed in light of all the effort and energy on Android, Symbian, and Windows? Only time will tell.

End Note:

1)  Cannistra oversees the company’s strategy to embed mobile WiMAX functionality into notebooks, netbooks and a wide range of consumer electronic devices. He is responsible for Clearwire’s strategic relationships with Intel, PC OEMs and numerous other companies that aim to make wireless broadband accessible to more people and devices.

Viodi View – 07/01/2009

Summer is here and the living is easy; at least I think that is how the song goes. Actually, summer is here and somehow I missed spring cleaning. Loose ends seem to be everywhere; web sites that are almost updated and articles not quite finished. With all of the clutter, my mind becomes blurred with visions of the way I picture things were supposed to be and the reality of what they became.

3G-HSPA, Mobile Linux and Open Source are the Big Winners in Intel-Nokia Technology Partnership by Alan Weissberger

The Intel-Nokia announcement of their joint push to work together to create a new breed of Mobile Internet Devices based on various versions of Open Source Linux, reminds me that what you see in press releases is not always what ends up being reality. Of course, we see this all of the time in the technology industry, but Alan Weissberger and several of his informed readers point out previous Intel announcements in the mobile space that haven’t quite led up to the vision of what was suggested. Another major thing missing in this announcement is Microsoft.  Click here to read Weissberger’s sharp analysis as well as astute commentary from his readers.

Connections at ConnectionsKurt Scherf of Parks Associates summarizes the 2009 Connections Conference

The Parks Associates conferences are good events for providing a glimpse of what might be in the consumer electronics’ world. In our ongoing video coverage from last month’s Connections conference event, Kurt Scherf, Vice President of Research for Parks Associates discusses how devices in the home are becoming more web-like and the on-going challenges of connecting these broadband devices.  Click here to view the video.

Via Licensing video interviewLicensing Pools with Via Licensing

I.P. (Intellectual Property, that is) rights owned by multiple parties are often one of the reasons standards take so long to move from concept to commercial reality.  In this video interview, Jason Johnson, of Via Licensing, a subsidiary of Dolby Laboratories, explains their newly formed partnership with the IEEE.  This partnership promises the expeditious creation of licensing pools, allowing standards to be commercialized much sooner than traditional approaches.  The upshot of this effort should be even faster innovation by consumer electronic companies to bring us new gizmos and widgets.  Click here to view the video.

What’s New with Vegas 9 Provegas 9 video reviewed by Roger Bindl

Matthew Brohn, Product Manager of Sony Creative Software, discusses some of the strengths of Vegas Pro video editing, and what’s new with Vegas 9… things like small footprint, direct editing of AVCHD and XDCAM, plus new effects. The video includes screen captures of real-time editing in Vegas.  Click here to view Roger’s review of Vegas 9 Pro.

People on the Move:

Congratulations to Nsight/Cellcom and for the award they received from the Femtoforum for, “Significant progress or commercial launch by a small carrier .” They won this award for their, “deployment of the world’s first IMS-based, CDMA femtocell network for consumers and enterprises.”   Rob Riordan of Cellcom was in London last week to accept the award.

As follow up to an article I wrote last year when I announced that I was involved with a stealth start-up, this is the official announcement that the referenced company is ZillionTV.

The Korner –  Loose Ends EverywhereViodiTV Revealed - the Video

Roger and I are always so busy producing content that often times the packaging around the content is somewhat unfinished. It is sort of like the last bit of molding on the remodel that just never gets installed; most guests won’t notice it, but, to the owner, will view it as an eyesore. We have a great deal of unfinished business on the Viodi View and ViodiTV web site, which may or may not be obvious to the visitor.

Despite the packaging, it is the content that matters. Roger recently put together a video to tell the story of ViodiTV. Roger is normally ruthless at cutting out extra content, but he found it difficult in this case, as we have had the good fortune to interview some really cool people and report on some really interesting stories over the past several years. This video is really a commercial for Roger’s talents, but it also provides the story of ViodiTV and our attempts to tell the stories of the Independent Telcos and their communities.  Click here to view the video.

3G-HSPA, Mobile Linux and Open Source are the Big Winners in Intel-Nokia Technology Partnership

Intel referred to it as "this year’s most significant collaboration in our respective industries." The Intel-Nokia strategic partnership will "align and shape the next generation of mobile computing." But it was very difficult to extract any tangible take always from the press conference announcing the partnership. That’s because no specific products were identified and no time frames were given to see the results of this highly acclaimed collaboration.
To a large extent, the press conference was a lot of hand waving and gesturing, without providing anything of substance that we might expect from such an important strategic relationship. This is the third time in the last decade that Intel and Nokia have announced a partnership, with the previous two attempts not producing much if anything at all. So the industry might have a right to be skeptical this time. Nonetheless, it certainly sounded exciting. 
Anand Chandrasekher, Sr VP and GM of Intel’s Ultra Mobility Group stated, "The leaders in both computing and communications are coming together to accelerate innovation while driving exciting new revenue opportunities. Intel and Nokia are joining forces to announce a long term strategic relationship that will align and shape the next generation of mobile computing."
We are all aware that smart phones and intelligent hand held devices contain powerful processors and need mobile broadband capability to unleash "the tremendous power and potential to reshape our lives." With many different wireless communications options, most of us expect that "the future will bring even more ways to be connected- a future full of different possibilities." Yet that kind of talk dominated the prepared remarks of Intel and Nokia during the conference. 
Nokia told us more of what we already know: "The Internet continues to evolve and touch every aspect of our daily lives. Today, there are more Internet users (at 1.6B) than there are fixed phone lines (at 1.3B). There are over 64B web sites exist and more are added every day. And the Internet continues to grow in every aspect. New applications will drive the need for more powerful compute engines and faster (mobile) broadband wireless access, Consumers looking for mobile devices to do more, e.g. sensors, new apps, new materials, new device design and form factors (e.g. netbooks MIDs). We need to extend computing platforms, build on common open platforms and explore new architectures."
Anand told us that Intel would continue to "relentlessly focus on driving down the cost and power requirements (of new devices), while delivering continuing performance improvements." Should we have expected something different? We were also told more of the obvious, "Mobile devices require high bandwidth- mobile broadband communications and ubiquitous Internet connectivity at a reasonable cost. Users should expect a rich experience, any time, anywhere. New and exciting services across a range of devices, including new ones the companies will be defining together."
So what’s really new? There are three aspects of the partnership, which is not limited to just hardware and Research and Development:
  1. Intel and Nokia will collaborate on several open source initiatives, most importantly Mobile Linux. Nokia pointed out that "Hardware and software are decoupled these days. Mobile Linux is an important part of the new converged mobile computing world." We would expect Intel and Nokia joint software development to be centered on two open source projects:
  • Moblin, originally an Intel project but now run by the Linux Foundation.
  • Maemo, a Nokia implementation created for an Internet tablet.
  1. Intel is licensing 3G HSPA modem technology from Nokia, complementing its own WiFi and WiMAX silicon. (Note that two years ago, Intel licensed an HSPA module from Nokia for use in notebooks. This technology transfer is intended for Intel to offer HSPA silicon for mobile hand held devices).
  1. Intel and Nokia have entered into "a long term strategic partnership to develop a new class of mobile computing devices." Those future mobile computing devices will be based on Intel architecture defined chip sets and will "leverage each company’s expertise." 
And what about Mobile WiMAX? Don’t expect anything from the partnership. In response to a question on further WiMAX co-development, Anand replied, " This announcement has no effect on WiMAX one way or another. We are still committed to it. In this announcement, we are expanding our wireless portfolio to be able to implement Nokia’s 3G HSPA technology."
–>This implies that Intel will no longer debunk HSPA technology in favor of Mobile WiMAX and suggest that network operators leapfrog 3G and move to Mobile WiMAX instead.
When a questioner pointed out that Nokia now had licensed 3G-HSPA to five different companies, Intel and Nokia responded as follows.
Intel: "3G HSPA technology has been licensed to build into future mobile offerings. No comments on products or timing. Nokia and Intel’s vision is very similar- bringing communications and computing together. This is not an exclusive agreement."
Nokia: "3G HSPA is what’s on the market today (implying Mobile Wimax is NOT really on the market). Nokia is licensing its 3G-HSPA-modem technology as widely as possible within the industry.”
When asked if Intel had made any other inroads in the mobile phone business (which the company has tried to crack for years, but has not succeeded), Anand replied, "Intel is not public on any wins in the mobile phone arena except for LG." Then when asked what type of LG device would be forthcoming, Anand would not comment on the specific LG device that will have "Intel inside." 
Author’s Note:  This was surprising, considering that Intel had previously touted the LG MID (with Ericsson HSPA module) as the highlight of this year’s Barcelona MWC.


The stonewalling continued in response to other very reasonable questions about partnership deliverables:
Question from Bloomberg News: “There have been a lot of announcements about visions of the future. Intel has tried to get into the mobile communication business for a number of years, yet they have not succeeded. There’s still a degree of skepticism until we know when the first Intel powered mobile device will be out there. Can you tell us?”
Intel: "We will work together on strategic technology collaboration which spans three areas: Intel Architecture defined chip sets for future mobile computing devices, mobile and MIMO collaboration to deliver a very rich software environment for applications and user experience, Intel licensing of Nokia’s 3G HSPA technology   No products announcements at this time- not for today’s discussion."
Question: "Do you expect the Atom family (Intel’s lower power micro-processors) or x86 family to be embedded in future mobile computing devices?"
Intel: "No comment on brands or usage."
Nokia: "Premature to say how we will apply the technology at this stage."
How will the Mobile Computing Industry be affected as a result of this partnership?
  1. Could these mobile devices, with open source operating systems like mobile Linux, cause MSFT to lose its software domination of the computing world? Could this mean the end of Wintel dominance of the computing industry?   Mobile Linux- one of the three focus areas for Intel-Nokia partnership – is a direct competitor of MSFT’s Windows Mobile. As people increasingly use mobile computing devices to do things that would have required a PC a few years ago, MSFT is likely to lose ground.   Mobile computing devices, e.g. smart phones, MIDs, all-in-one gadgets, etc are already replacing a lot of things we do today on PCs. This trend will likely accelerate as mobile computing replaces desktop computing.  
  2. Does this announcement negatively impact Mobile WiMAX, which already has been severely criticized for the lack of mobile devices with native mode air interfaces? After all the Intel talk about WiMAX MIDs, we are still waiting for those devices to hit the market in a big way. Will "the Internet in your pocket," be based on 3G-HSPA, rather than Mobile WiMAX?
An anonymous Intel employee provided his read on the partnership:
"This announcement does not change any of Intel’s plans on WiMAX which are solid going forward. Intel has not been a major player in Smart Phones/MIDs and we want to get into that space with the Intel Atom® Processor so this one part of this strategy. Also most smart phones shipping today at least have 2G/2.5G and many also 3G. So this licensing deal help fill a gap in our wireless technology portfolio. 
It also allows us to provide WiMAX solutions to Nokia once more networks get deployed and they want their mobile devices to have WiMAX support as well. So by no means does this negatively affect our WiMAX strategy. It only opens new doors for us with a large customer like Nokia."
  1. When will the new mobile computing devices hit the market? They will need to come quickly, if they are to compete with all the new smart phones from Apple, RIM, and Palm. We hear there will also be MIDs coming soon from Samsung and various Taiwanese companies. Previous Intel – Nokia partnerships, e.g. HSPA modules for notebooks, have not been successful so the industry is skeptical that this one will succeed. We would expect to see Intel-Nokia mobile computing devices on the market in less than one year and perhaps as early as this Christmas.
Intel and Nokia Announce Strategic Relationship to Shape Next Era of Mobile Computing Innovation
Intel makes stab in the dark with Nokia deal 

Intel- Nokia Partnership Facing Market Challenges

Gary Reback: US Government Must Enforce Antitrust Laws to Encourage Innovation

On the need for government intervention in the free market to protect competition and encourage innovation

Gary Reback is one of the nation’s most prominent antitrust attorneys, best known for spearheading the efforts that led to the federal lawsuit against Microsoft.   Gary spoke to an attentive and eager audience on May 14th in Santa Clara, CA.   The Commonwealth Club and Yale Club of Silicon Valley sponsored his enlightening and provocative talk. Reback’s main message was that the government l’aissez faire policies, so strongly promoted by University of Chicago economists, have gone way too far. As a counter-weight, he says we need more government oversight of the private sector along with more vigilant anti-trust enforcement.
To set the stage for the current recessionary economy, Gary began by chronicling the history of the U.S. antitrust movement. From its beginnings in the 1870s (a time when big business controlled the railroads), through Teddy and Franklin D. Roosevelt, Thurmond Arnold and others, there has been an ebb and flow of power and control between the federal government in Washington and big business (e.g. monopolies) or Wall Street investment firms. Starting about thirty years ago, conservatives forced an overhaul of competition policy that has loosened business rules for everything from selling products to buying competitors. In the free market era of the 1990s, big business and investment banks certainly had the upper hand. To a large extent, that is what has caused the global financial meltdown and enabled companies like Microsoft, Intel, Cisco, Oracle, and Google to become so powerful. We were surprised to learn that expensive medical care and non-generic prescription drugs are the result of a lack of anti-trust enforcement.
Mr Reback firmly believes that in a high-tech world, U.S. government “hands off” policies actually slow innovation, hurt consumers, and entrench big companies at the expense of entrepreneurs.   In particular, Gary calls for increased government scrutiny of high tech firms monopolistic practices. He argues that monopolies have the power to raise prices by restricting output, supply and competition. As a result, the economy weakens, unemployment increases, and innovation is pressured.
We take it for granted now, but many of the advances in semiconductors and software were the result of a few dominant lawsuits against big companies. 
In the late 1950’s, AT&T was forced by the U.S. government to license the transistor. William Shockley, one of the co-inventors of the transistor, licensed it from AT&T to form Shockley Labs, which later begat Fairchild Semiconductor, which in turn begat Intel, AMD, and National Semiconductor. The early years of the semiconductor industry in Santa Clara Valley (it was not called Silicon Valley till the mid 1970s) was therefore, a direct result of the lawsuit against AT&T. For more on those early years, please see the article by this author:
In the early 1970s, concerned about possible anti-trust legislation, IBM was forced to unbundle software from hardware. This created a whole new independent software industry, which had not existed before. Software had previously been bundled with mainframes and minicomputers made by the same computer manufacturer.
Merger enforcement is perhaps the biggest business issue of our time, according to Reback. What do we do with companies too big to fail? How about Citibank Group, for example? If it had not been for the repeal of the Glass Steagall act in 1999, Citi would not have been able to acquire Smith Barney, Solomon Brothers and other investment firms. Hence they would not have gotten too big to fail.   Better to have government carefully scrutinize the mergers and acquisitions and/or break up large companies before they become too big to fail! One has to wonder if certain tech companies, like Oracle and Cisco have become too dominant in their industry or even too big to fail because of acquisitions that occurred without anti-trust scrutiny. For example, Oracle has done over 40 mergers after it acquired People Soft and now has proposed to acquire Sun Microsystems.
Is Google the next Microsoft? Yes, in terms of its dominance over web search software, in comparison to Microsoft control over desktop and notebook PC software. No, in at least two other important ways: 
  1. Google created technology that people liked and it worked well. 
  2. The company was also more customer friendly with a more congenial corporate culture and image.
Author’s Note: the next battle between these software titans will be in mobile OS market- Android platform from Google vs Windows Mobile from Microsoft.
Gary believes that Google’s big search competition will come from social networking sites (e.g. Twitter and Facebook), rather then from traditional search engines from Yahoo or Microsoft. He also noted that potential anti-trust action was enough for Google to call off its plans to put adverts on Yahoo’s search result pages.
Was the European Union’s (EU) huge fine against Intel Corp justified? Just one day before this talk- on May 13th– EU regulators slapped a record 1.06 billion euro ($1.45 billion) fine on Intel for antitrust violations and ordered it to halt illegal efforts to squeeze out arch-foe AMD. This fine was levied after an 8 year EU investigation of the company. "Intel has harmed millions of European consumers by deliberately acting to keep competitors out of the market for computer chips for many years," EU Competition Commissioner Neelie Kroes told a news conference.
Should Intel have known better to refrain from engaging in unfair trade practices? Most definitely yes, according to Gary. Intel was actually a U.S. government witness in the huge anti-trust suit against Microsoft in 1998. The plaintiffs (US Department of Justice and 20 states) alleged that Microsoft abused monopoly power on Intel-based PCs in its handling of operating system and web browser sales. The issue central to the case was whether Microsoft was allowed to bundle its flagship Internet Explorer web browser software with its Microsoft Windows operating system. Bundling them together is alleged to have been responsible for Microsoft’s victory in the browser wars, especially over arch rival Netscape (which seemed by this author to be a superior web browser). Didn’t Intel learn anything from the trial and the verdict against Microsoft regarding unfair competitive practices? Gary response, "Of course, Intel says it is innocent of the charges and never broke the law, so perhaps the company will be exonerated after the EU Commission decision is reviewed by the European courts."
Opinion: We suggest the reader to ask the question to an Intel executive or lawyer.
In summing up, Reback opined that “anti-trust action failures” in the health care and banking industries have contributed to unreasonably high medical costs and a financial meltdown. Meanwhile, heightened scrutiny over acquisitions (e.g. Oracle’s) would result in a stronger U.S. economy by encouraging more competition and invigorating innovation and the start up culture. 
Bio: Gary Reback is one of the nation’s most prominent antitrust attorneys. He has been named one of the “100 Most Influential Lawyers” in America by the National Law Journal and is quoted regularly by major media. His book Free the Market! is a memoir of Reback’s titanic legal battles—involving top companies such as Apple, Microsoft, IBM, Oracle, and AT&T—and a persuasive argument for measured government intervention in the free market to foster competition.   Gary is currently of counsel with Carr & Ferrell LLP. He is a very friendly and easy to get a long with person, in this author’s opinion.

WiMAX in smart meter sets the stage for "open standards" based smart grids

San Francisco start up software company Grid Net, is using WiMAX technology in its wireless smart meter instrument. The meter, being built by GE, uses Intel’s WiMAX chip and Grid Net software. It is said to be one of the first truly "open-standards" based approaches to building a meter.

Grid Net’s meter went on sale to select utility customers this March, and though Bell declined to state the current price, said that within a year and a half he expects the price to drop low enough to undercut any of the meters on the market that use proprietary technology. Bell says he has four major deals in the works and contends that his smart meter will eventually be cheaper than proprietary systems on the market (this is a key benefit of "open-standards").

Because WiMAX operates over licensed wireless spectrum, Grid Net founder Ray Bell claims it’s far more reliable and secure than unlicensed wireless networks (e.g. WiFi) — a particularly important feature for smart grid deployments. The meters could use national WiMAX networks (e.g. Clearwire or from regional carriers) or WiMAX networks that would be built and owned by a utility.

GigaOm states, "but if the smart grid really will follow the lessons of the Internet, open standards will be a key driver."

The use of WiMAX for grids is not a new concept.  In January 2005, we wrote that:


The basic premise was that fixed WiMAX could be effectively used to extend a carrier’s long haul network for access to a grid computer network.  We are now seeing WiMAX used in emerging smart power grid networks, but the concept is the same.  We think WiMAX has a lot of potential and promise for interconnecting PCs and meters, and other instruments over smart power grids.

WCA Panel Session: Femtocells and their consequences for Mobile Broadband Technologies


Femtocells are low power 3G/4G cellular base stations which can be thought of as wireless Access Points (AP’s). Instead of a WiFi Access Point we’re all familiar with, a femtocell could be a UMTS, CDMA/EVDO, LTE, or Mobile WiMAX Access Point, depending on the underlying mobile network technology. The bi-directional voice and data traffic is taken off the respective wireless network and placed on the Broadband Internet connection at the home or office. Femtocell suppliers claim they’ll be capable of communicating at 3G (HSDPA) and 4G (LTE, mobile WiMAX) speeds, depending on the bandwidth of the broadband Internet connection (which could be a bottleneck).
Femtocells will provide mobile handset and notebook PC coverage in a home or business with typically 2- 4 simultaneously active users. Picocells are larger APs designed for enterprise use, with higher output power, to handle more simultaneous users. No changes are required to the handset, notebook PC, or other mobile devices, [the SIM card in a mobile device might be programmed to authenticate the user].   The residential and enterprise femtocell markets have very different requirements and operators will likely target one of those market segments. Some carriers see the need for enterprise femtocells creating a bigger market opportunity than residential. With few exceptions (e.g. T- Mobile USA), femtocells are preferred to WiFi APs by mobile carriers.
Definition: Here is a link to the Wikipedia definition:
The Femto Forum is a not-for-profit membership organization founded in 2007, to promote femtocell deployment worldwide. They have published results of a comprehensive study project, which shows a positive femtocell business case:
Here is another useful web site:
On March 17th 2009 the WCA Mobile SIG sponsored a panel on Femtocell technology and business models. The participants were as follows:
Moderator and Presenter: Stefan Scheinert, Principal, Scheinert Telecom
— Peter Walther, Product Manager, mimoOn
— Dror Nahumi, Partner, Norwest Venture Partners
— Behzad Mohebbi, CTO, Nextivity
— Tom McQuade, VP NA Sales, picoChip
— Michelle Pampin, Wireless Backhaul Specialist
One primary goal of femtocells is to take traffic off the mobile network, which operators have paid a substantial spectrum license fee to own and use for mobile services. With more and more voice and data traffic originating or terminating in homes or offices, this would free up spectrum for users on the move. It could permit more mobile users per cell site, OR smaller and cheaper base stations (with less output power), OR higher speeds per mobile user. However, this is all dependent on the broadband Internet connection and ISP policy regarding femtocell traffic. For example, a DSL connection would probably not provide enough bandwidth to carry traffic from multiple home or enterprise femtocell users. If the ISP is not the same as the mobile provider, it could block the traffic or at least throttle it back (e.g. Comcast and Bit Torrent). Why should the ISP give a free ride for traffic that would otherwise be on the mobile operator’s network?
Are femtocells the answer? There are several competing technologies, including:
·                    Unlicensed Mobile Access or UMA (WiFi based)- As noted earlier in this article, T Mobile is using this approach to provide VoIP over WiFi in the home, with GSM cellular voice outside. Users get dual mode handsets and calls automatically switch from one wireless medium to the other. But we suspect this might be a stop-gap measure until femtocells are ready for commercial deployment (see below).
·                    Macro Base Stations– with more output power: Seimens, Nokia, Ericsson are expected to build these higher throughput Base Stations (with up to 300 M bit/sec aggregate data rate).
·                    Repeaters– favored over femtocells when there is no high speed backhaul available, or the operators do not want to use femtocells. It was stated that LTE will likely use repeaters, rather then femtocells, due to higher speeds not supported by the broadband Internet connection. Vodafone, which has also conducted femtocell trials in a number of European countries, is said to be expanding its trials of repeaters in several European locations, although the company has admitted it would continue to evaluate both femtocells and repeaters.
Market Assessment: Norwest Venture Partners believes these different alternatives have created enough confusion to delay the femtocell market. As a result, the market has been slow to take off and difficult to justify. And even if the market for femtocells picks up, it might be difficult for a start-up to make money.
Voice over LTE: The current 3GPP release 8 specifies LTE as an all IP network, but does not include support for a voice codec. Most likely, voice will be offered via GSM or CDMA codecs, or a Skype like VoIP over LTE service. 3GPP Release 9 or 10 will specify voice codecs for use with LTE.
Here are a few critical issues/ checklist for femtocell deployment
(Source: Stefan Scheinert):
·                    Femtocell AP Box cost should be < $100
·                    Location awareness required, because Femtocells use licensed spectrum
·                    Macro network interference (from other femtocells and outside mobile network cells) requires dynamic management, i.e. an interference detector/sniffer to measure power strength of neighboring cells.
·                    Security and integrity of femtocell traffic over the Internet using the femtocell backhaul or broadband Internet connection.
·                    Plug and play with auto configuration on power up.
Author’s Opinion: Another issue (similar to WiFi free-loading) is how to ensure individual femtocells are only accessible to the homeowner or business paying the bills. Some type of automatic authentication will be required prior to use. And there is a national security issue as well- how operators can comply with regulations that allow for the interception of cell phone calls by law enforcement agencies?
Deployment of Femtocells
There are a large number of operators participating in the Femto Forum, which augurs for actual deployment in the near future. Picochip stated that they have licensed their femtocell technology to IP Access, which is working with Cisco to get femtocells deployed at AT&T. Cisco is contributing Self Organizing Network software to this femtocell initiative.   Three cities trials were said to take place in 3Q 09 with commercial deployment scheduled for 4Q09. In checking with my trusted colleague at AT&T, I was told that currently there are internal femtocell trials within the company, but nothing else has been announced.
We hear that there are many carrier field trials of femtocells, but no results have been announced. Many different variants of femtocell technologies are being trialed, according to our privileged sources. For example, T-Mobile (which has deployed VoIP over WiFi in the U.S.) has completed several trials with femtocells, and invested in the UK-based femtocell manufacturer Ubiquisys. It is said that the company plans to launch a commercial service using the technology by mid-year in Germany.
“Femtocell rollouts to date have been limited, controlled ones,” said Aditya Kaul, an ABI Research senior analyst. “Shipments at the end of 2008 were in the few hundred thousands, and at the end of 2009 should climb towards a million but will fall short.”   ABI Research expects that 2010 will see shipments climbing well above a million units. Indeed, vendors are gearing up for a big push, Kaul said. One of the industry’s main silicon suppliers, PicoChip recently announced a multi-million dollar injection of funding. Kaul said he expects that the funding will “probably be geared toward a ramp-up and that there is “a lot of similar activity behind the scenes, and new partnerships which point to preparation for a major market expansion.”
Late in 2009 or early in 2010, ABI Research said it expects an announcement of a multi-city commercial femtocell deployment by one of the major mobile operators, which may encourage other operators to follow suite.   Until now, large-scale femtocell deployments have only been simulated in computer models: real-world rollouts could pose challenges.
Price is one: ABI Research believes that although femtocell business models could be enabled at various price points, low-cost femtocells (under $100) are essential to bridge the gap between niche market and mass-market deployments.
Nonetheless, “These challenges are all valid, but none of them are show-stoppers – there’s no ‘elephant in the room’ that will pose a major obstacle to large-scale deployment,” according to Kaul. For more information, please see:
Recession Slows Femtocell Deployments, but Only Temporarily, According to ABI Research
There was some uncertainty expressed by the panelists, as to which network operator would actually deploy and maintain the femtocells. We think it will be the mobile operator that is offloading network traffic and providing better indoor reception to its customers. But Michelle Pampin, Wireless Backhaul Specialist, stated that it would be the provider of the broadband Internet service (who controls the QoS for the femtocell traffic that’s backhauled to the public Internet). Panel Moderator StefanScheinert told me privately that Michelle was not correct, i.e. that the mobile operator would own and sell the femtocells to end users/ enterprises.
Here’s a chart of Femtocell Deployment status in Europe:
Any WiMAX Femtocells?
There has been a lot of speculation about Comcast providing femtocells for its mobile WiMAX deployment (reseller agreement in Portland, OR announced and several cities have been targeted for mobile WiMAX build-outs. But at this time, there are more questions than answers.
Comcast initially said that femtocells would be part of its WiMAX strategy, but nothing was said about it when its Portland,OR WiMAX service was announced last week (reselling Clearwire’s CLEAR).   Comcast’s Sr VP for wireless and technology, Dave Williams, said in June 2008 that a key element of the Clearwire investment and partnership was to reserve 5 MHz of spectrum for WiMax femtocell deployments. That spectrum would also be available for use by any of the Clearwire consortium members, which includes Comcast, fellow cable MSOs Bright House Networks and Time Warner Cable Inc. as well as Google and Intel Corp.
“We’ll be pushing WiMax femtocells because we have a good customer base in the home — we sell HDTV, VOIP, and high-speed Internet connectivity. We want to take that experience in the home and add mobility,” Williams told Unstrung at the time. But Comcast has not said anything more about WiMAX femtocells- at least not publicly. For more on this topic, please refer to:
Conclusions and Companies to Watch:
As with so many new technologies, the jury is still out on femtocells especially for Mobile WiMAX, which has yet to establish a critical mass of subscribers. We are most concerned about the broadband Internet connection being a potential bottleneck (especially if it is DSL) and the broadband ISP blocking, metering or restricting femtocell traffic on its network. Stefan does not think the ISP will actually block femtocell traffic (note that the FCC fined Comcast for doing this). So there is a likely to be some contractual agreement between the mobile operator supplying the femtocells and the broadband ISP.
We would watch PicoChip, which appears to have leading edge femtocell technology. Please see this overview and ecosystem report:
Percello is an Israeli start-up femtocell manufacturer to also keep an eye on. They are designing a low-cost, high performance femtocell 3G W-CDMA baseband processor chip. Please refer to:

Google (and Intel) write down Clearwire investment; Google optimistic about Android platform

Google’s fourth quarter results were hurt by write downs related to investments in AOL and Clearwire. The Internet search leader also said it remained confident about the progress of its Android platform for mobile handsets.

Google reported a net income of $382 million, down 68 percent from the $1.21 billion in the year-ago period, but that was primarily due to write downs of its investments in AOL and Clearwire by $726 million and $355 million, respectively. Google CEO Eric Schmidt stated, "Both deals (Clearwire and AOL) made sense to us then and make sense to us now, and continue to be a strategic part of our overall business philosophy,"

Google, along with Intel, Comcast, Time Warner, Brighthouse and others invested $3.2 billion in the new Clearwire. All of these Clearwire investors face huge losses.

Intel recently took close to a $1B charge on its Clearwire investment (Intel’s profits were down 90% in the most recent quarter). Clearwire’s stock has dropped more than 70 percent since the deal was announced in May, resulting in a $950 million “impairment charge” for Intel. That comes less than a month after Intel Capital’s head honcho- Arvind Sodhani – stated he was excited, rather than worried, to have sunk so much of Intel’s portfolio into Clearwire. Of course, it must have been obvious even then that the investment would end the year with a big loss.

Google also expressed support for its Android platform, even though it only has limited support among U.S. wireless carriers. T-Mobile USA is the only carrier to date that supports an Android-based handset, the HTC-made G1. Sprint will reportedly begin offering an Android handset in the first half of 2009.

Note that many pundits think Google will finance a company to build a WiMAX handset, due to their investment in Clearwire. We disagree.  In our opinion, the Clearwire stake was a strategic investment to enable more people to access Google’s ads, maps, and other content.

Do you think the future Sprint Android device will have embedded WiMAX capability to support CLEAR?  Note that Clearwire already stated that VoIP would not be introduced for CLEAR anytime soon.

Google Very Optimistic About Android

"Well, we think the Android numbers will start to speak for themselves, particularly this year, now that we’ve got some additional partners signed up to get to the next generation from a hardware perspective on the platform," said Jonathan Rosenberg, Google’s senior VP for product management. "So we are very optimistic about Android."

Here is what one blogger thinks about Google’s business going forward:

Google keeps its one-trick pony healthy

What do you think about Google’s Clearwire investment and prospect for Android handsets- with and without WiMAX capability?

Where are the WiMAX MIDs?

Did you notice the absence of any major WiMAX device announcements at CES? With all the talk by Sprint and Clearwire about mobile WiMAX at the show, we were expecting a raft of WiMAX capable Mobile Internet Devices (MIDs) to be announced at CES. We were disappointed.

After conversing with attendees and scouring the web, we only saw one WiMAX MID announced at CES- a blog post about Yukyung’s Viliv "X70 AtoM Communications MID."

The company is based in Korea so we don’t know if the MID is only for the Korean WiBro market or will be sold worldwide. According to a Yukyung rep, the Viliv devices should be available in Spring 2009, priced between $500 and $700. Like all WiMAX devices, they will first have to be certified on each network they connect to.

For more info please visit:

Please see my summary of our local (SC Valley) Jan 08 IEEE ComSoc meeting presentation: Internet in Your Pocket, by Rama Shukla of Intel

Hence, my keen interest in why there have not been more MIDs announced – one year later! Without a large choice of MIDs, netbooks and notebooks with embedded WiMAX, I don’t see how there can be a market for mobile WiMAX.

In sharp contrast, there are MIDs for 3G+ networks that may be coming to market this year. At CES, Qualcomm’s demonstrated the Android software platform running on its Snapdragon chipset, which is targeted at non WiMAX MIDs and other devices that go beyond the conventional handset. To the best of our knowledge, none of the Android based MIDs or phones in the works will support mobile WiMAX. They are all 3G/ 3.5G based. Qualcomm said the work is geared to larger MIDs and web tablets, having already been an early adopter of Android on its mainstream smartphone chipsets – the first Android phone, the T-Mobile/HTC G1, runs on the Qualcomm MSM2701A silicon.

Opinion: Snapdragon could be stiff competition for Intel’s AToM processor, which is also targeted at MIDs (with built-in WiMAX adapter). The MID introduced in 2009 may be mostly 3G/ 3.5G since those networks are much more pervasive than mobile WiMAX. Alternatively, the entire MID market may never come to pass- with equivalent functionality built into smart phones and netbooks. We are looking forward to seeing WiMAX enabled netbooks by the end of 1Q09.

Update on Mobile WiMAX- is it really mobile (or only for wireless fixed line access)?

Over one year ago, I wrote an article titled: Will the real mobile WiMAX please stand up:

The claim then was that IEEE 802.16e (AKA Mobile WiMAX) would be primarily used for fixed line BWA. Not much has changed in the last 18 months. Recently, there’ve been some encouraging announcements from Clearwire, Digital Bridge , and several overseas wireless service providers.  However, the overwhelming majority of WiMAX deployments continue to be for fixed BWA.  True mobile WiMAX service has been hampered due to several factors: a scarcity of native mode WiMAX devices, developing countries urgent need for fixed BWA, and the uncertainty of mobile VoIP (over mobile WiMAX) availability and roaming.

Where are the Native Mode WiMAX devices and MIDs?

The WMAX Forum says there are 480 WiMAX devices in development – by over 80 vendors. But we know of only one native mode WiMAX device – the Nokia 810 Internet Tablet Computer. We’ve heard a lot from INTEL about embedded WiMAX in Mobile Internet Devices (MIDs), but we haven’t seen any MIDs hit the market yet. While several notebook PC’s with embedded WiMAX chip sets have been announced, we don’t know of any available. Instead, dongle attachments are now needed for notebooks to access a WiMAX network.

Intel is now shipping its first-ever combined WiMAX/Wi-Fi module, formerly code-named "Echo Peak," as part of the Intel® WiMAX/Wi-Fi Link 5050 Series that is an optional feature for Intel Centrino 2-based laptops. Several leading PC OEMs have announced plans to deliver Intel® Centrino® 2-based notebooks featuring the Intel® Core?? processor with WiMAX including Acer, Asus, Lenovo and Toshiba. Beginning today, WiMAX-based Intel notebooks can be found online at and Dell, Panasonic, Samsung and Sony also plan to support WiMAX in their laptops starting in 2009.

Perhaps the scarcity of true mobile WiMAX devices is power consumption of Intel’s AToM processor and power amps.The general opinion is that the power added efficiency (PAE) of the power amps used in Mobile WiMAX terminals is far too low. Power amps for mobile phones, for example, offer PAEs of 40% to 45% for wideband code division multiple access (W-CDMA), and 50% to 55% for Global System for Mobile Communication (GSM). For Mobile WiMAX, though, the PAE is only 10% to 20%.   For more on this issue, please see:    Can WiMAX Really Go Mobile?

Where is Mobile WiMAX being deployed?

The U.S. carrier that is offering true mobile WiMAX is SPRINT (which is now part of the "new Clearwire") with its Xohm network- operational in Baltimore since October 3rd. The only handheld device we know of that’s certified to connect to Sprint’s Xohm WiMAX network is the Nokia 810 tablet PC noted above.

For more information, please check:

The "new Clearwire” recently announced that, while its company name will remain Clearwire, its new mobile WiMAX services will be branded Clear™. The Clear™ brand will apply to all new mobile WiMAX services to be offered by Clearwire in the U.S. and will be phased in to those markets where Clearwire offers pre-WiMAX services, as these existing markets are upgraded to mobile WiMAX technology.

CEO Ben Wolff said the company plans to upgrade most of the existing 46 Clearwire markets to WiMAX next year. He added that Clearwire’s cable partners (e.g. Time Warner) are expected to offer wireless services under their own brands using the Clearwire network.  Clearwire’s mobile WiMAX roll outs, together with MSO reselling, significantly increases the chances that mobile WiMAX will emerge alongside LTE (Long Term Evolution) as one of the technologies that will power the next generation of high-speed wireless data networks.

At the November 2008 WCAI Symposium, Digital Bridge Communications stated they plan to offer mobile WiMAX (along with VoIP) in 15 U.S. markets. Handoffs to other carriers along with full mobility will be inherent in that network, which we need to watch closely to see if there are many dropped calls or Internet sessions. The network, known as BridgeMAXX, operates on licensed 2.3GHz frequency and is said to be very secure.

What about the new 3.65 GHz networks in the U.S? All that were described at a WCAI session use the fixed line version of WiMAX (IEEE 802.16d or -2005). The 3.65 GHz band uses a non-exclusive or "light licensing" regime to promote rapid deployment of wireless broadband systems to rural and underserved areas.

What’s the status of true mobile WiMAX in other countries?

Korea Telecom (KT) is offering a form of mobile WiMAX as part of its WiBro service in Korea. The relationship between the two is detailed in this whitepaper:

Just last week, we read that telco YTLE is collaborating with Cisco to launch a mobile WiMAX Network in Malaysia. YTLE plans to deliver affordable WiMAX packages with fast, wide coverage and always-connected mobile internet services that will cater to and complement the personal and business communication needs of consumers as well as the enterprise community.

Other overseas carriers planning to offer true mobile WiMAX are: UQ (part of KDDI) in Japan, Scartel and Comstar in Russia. However, Lee Sparkman, principal at Entorta (another Russian Broadband Wireless carrier) stated their market was exclusively fixed wireless broadband. With a take rate of only 15%, BWA is cost effective in developing countries, while a DSL or cable build out couldn’t be justified.

Esme Vos of has posted a map of cities around the world that have "mobile WiMAX" service. She writes, "By “mobile” I mean you can connect to the WiMAX network either via a USB dongle or a WiMAX-enabled device (not many of those). You can view the map on:

The pain of finding mobile WiMAX networksI had no idea it would take so long to assemble a paltry list of cities with mobile WiMAX service. I went to several WiMAX websites to find WiMAX networks and make a selection. The problem is that although many of them indicate that there is a WiMAX network based on 802.16e (”mobile” WiMAX), in reality, you need a stand-alone modem to gain access to the network – not useful for travelers. Most of the sources I used are unreliable: in certain cities they say that the service is live, but when you go to the provider’s site, you discover it’s not.It took me half a day to cull through dozens of listings of “mobile” WiMAX service providers to come up with this limited list, which is not complete given that new providers will be launching service in the coming months. So think of the map as a movable target."   The full article can be read at:


WiMAX in India Revisited

While WiMAX in India is expected to be a reasonably good market, we know of no Indian carrier planning to deploy a true mobile WiMAX service. Rather then a 4G competitor, WiMAX in India is seen as a wireless fixed line replacement or as a substitute for DSL/cable service. It is now operational from Metamax, Pronto Networks, Reliance Group, and Sify in the 3.3G and 5.8G Hz bands. There will be more to come next year after the long delayed auction. Watch what Bharti-Airtel does. They are the largest mobile operator in India. In a WCAI session titled: Blueprint for Success in Emerging Markets, Farid Arifuddin, Managing Director, Vavasi stated his company planned to deploy a true mobile broadband wireless network in India based on a proprietary technology- not mobile WiMAX or LTE. For more on WiMAX in India:

In answer to a query I had on true mobile WiMAX in India, here is a comment by Amitabh on November 28, 2008:

“The frequency auctions and Policy will permit Mobile WiMAX. However it is a question of how far the industry is ready. The past deployments are not necessarily a pointer as the Certification of profiles and availability of equipment is very recent. “

We conclude the jury is still out on true WiMAX mobility for India

Predictions for the number of mobile WiMAX subscribers

In a WCAI session titled: Anywhere Network Synergies in the Evolution toward the 4G Technology Ecosystem, Berge Ayvazian, Chief Strategy Officer, Yankee Group asked how many true mobile WiMAX users there’d be by 2012. One audience member called out "ZERO!" Mo Shakouri, VP Marketing for Alvarion said the number would be somewhere between 0 and 100M. Berge later told me privately he expected no more than 30M mobile WiMAX subscribers by 2012.

What do you think the number will be and what countries will support true mobile WiMAX vs fixed line or nomadic configurations?