The Tsunami is Coming: Mobile Operators, Brace Yourselves…

Why the use of the word Tsunami? Normally, this is a very scare phenomenon at sea. On land, we are faced with Thunderstorms, hail, wind, rain and so on. So, why ‘Tsunami’ in context of mobile broadband?  

The term 'Tsunami' comes from the Japanese, meaning "harbor" (tsu, 津) and "wave" (nami, 波). Source: Wikipedia. In the Tamil language, "Aazhi Peralai means “disastrous wave.” 

What mobile operators are faced with, for the next 5 years, is a series of waves when it comes to the increasing number of 3G subscribers, 3G enabled devices and laptops, application use and the enablement of video casting and photo sharing within social networks.

The first wave has already hit. Let’s call it the “Bold iPhone” wave of 2009

Last year, 30% of US consumers who purchased Apple iPhone 3G in the summer of 2008 switched from other mobile carriers to join AT&T. People want broadband and will make use of every bit in the network. In August last year, there were over 100 million 3G subscriptions out of a total of mobile subscriptions of 910.8 million in Europe (Source: Informa). It’s safe to predict that 3G subscriptions will reach 175 million by end of 2009 (iPhones , Blackberry BOLD, Android and other 3G phones).

Personally, I think the EO 2009 number will be closer to 300 million. According to The NPD Group, consumer sales of smartphones to US consumers represented 23% of all handset sales in Q4 2008 compared to just 12% in Q4 2007. By 2013, there will be 4.1 Billion subscriptions globally (source ITU). Of which 67% are 3G/3G+ capable (source: ABI).  That’s a staggering 2.7 million subscribers generating 80% of all global mobile traffic. 

Why is this scary? “A single high-end phone like the iPhone/Blackberry generates more data traffic than 30 basic-feature cell phones. A laptop aircard generates more data traffic than 450 basic-feature cell phones.” Source: Cisco Visual Networking Index (VNI) Jan 29, 2009:

Last week, it was widely reported that AT&T had network capacity issues due to heavy 3G usage in a central metropolitan location. We are talking about a few hundred to thousand subscribers congregating (imagine 60,000 people at Football game all using 3G to communicate and share video, photos, etc). Om Malik (Gigaom) predicted the impact on networks over a year ago. His March 16 post covered the AT&T network issue (See http://tinyurl.com/cn9pt8 )

The second wave: “the social broadband wave of 2010”

By 2011, we will most likely have over 1B 3G subscribers actively using multimedia to send/receive/share. In addition, 30-40% of the subscribers will watch live TV throughout the day, using mobile and/or Wi-Fi connectivity based on their location. This is reality today for some. Cable and Satellite TV can accessed vis phones and laptops (via DVRs, Slingbox and other services)

The third wave: “the TV/video wave of 2012”

Now, what happens when we have 2.7 Billion subscribers generating 2 Exabytes of usage, per month, by 2013?  We are talking about traffic doubling every year, a 66x increase per Cisco’s VNI where video accounts for 64% of all mobile traffic. http://tinyurl.com/b9berc

By 2013, the networks will crash unless something drastic happens. So what can be done?

• Why can’t we just add more cell sites?
  
• Why can’t we just add more spectrum?
  
• Why cant’s we just limit the # of 3G subscribers?
  
• Why’ can’t we just add Wi-Fi access points everywhere?
  
• Why can’t we just add Femto access points everywhere?
  

The answer is ‘YES” to all of the above. But, there are restrictions.

• Cell site costs can be as much as $500,000 in some cases. Even though AT&T is planning on spending $17 Billion on Capex this year, there are limits to how much an operator can afford to spend on network Capex, year after year. This is also why more and more operators are outsourcing management of their networks to Ericsson, Lucent, Nokia-Siemens, Huawei, etc – to reduce Capex. Of course, Opex increases but at least cost of equipment goes down. The net result is that the operators can demand better coverage and capacity from their service partners as demand for capacity increase from their subscriber base.
  
• More spectrum is needed. Operators want more uniform global bands to  improve roaming across continents and to help reduce device BOM and price in the store.. Likely candidates for uniform bands for 2012 and beyond include 700MHz, 900MHz, 2.5 and 2.6GHz. Members of 3GPP and ITU are lobbying for more Spectrum from US and European regulators. From the time spectrum is awarded or freed up, it takes anywhere from 2-5 years to deploy sites and enable devices to work across the band in question.
  
• You cannot limit the number of 3G subscribers, so take that option off the table (consumers have options with regard to operators)
  
• Wi-Fi with 802.11n will help offload usage at home and within businesses much like Femto access points deployed strategically in suburban areas and within the enterprise. Today’s Femto products are yet not ready. Femto access points need more intelligence, interference management and scalability to help offload capacity from the Macro network to improve overall subscriber experience as they roam across the land. By 2010-11, we should expect solutions to these challenges and Femto, WiFi and Macro co-deployments will help add capacity in “3G and 4G hot zones.”
  

So when will the Tsunami hit?  Well, it already did with a small wave caused by iPhone and Bold subscribers. If predictions are only 50% correct, we will face significant mobile network issues as early as 2010.  If Capex money is available, 2010-12 could be good years for infrastructure vendors and mobile managed services providers.

Twitter/mobileinsider

FEMTO Low-Price Market Penetration Strategy: a race to the bottom?

This week was filled with more doom and gloom predictions in the networking industry. In the midst of the forest filled with bears, AT&T made the statement that Capex is not down but slightly up as compared to 2008. Also bucking the trend, we heard supporting (positive) statements from IBM, Cisco, Google and Juniper as well as Vodafone. These companies are investing in people, companies and the future. So, overall, there are good news among a flurry of negative earnings and the Madoff scandal.

Some (perceived) sad news for the Femto players when RadioFrame finally confirmed that they will not be a player in the Femto AP market (rumors started prior to MWC in Barcelona). They will focus on the ASIC side of the business. But, is this bad news or good news?  Maybe RadioFrame decided that competing for business with basic access point BOM well above $200 and operator requirements for pricing at $200 going to $100 over the next 2 years -- was not a good idea. Can we fault that logic?

"We are trying to be prudent with our money in these tough economic times and we have pared back to a core set of businesses that will take advantage of our unique market position while maximizing our use of funds." - CEO Jeff Brown said in an email reply to Unstrung earlier this week.

I personally applaud them for making the tough choice. But, does that mean that the other players are running off a cliff? With enough funding and volume, a low-price, market penetration strategy can pay off (especially if your company name is Huawei and you have the entire China government supporting you). The dilemma facing Ubiquisys and ip.access (and now Airvana who is betting the future eon Femto) is that they are the first into a potentially enormous market that has enormous risks. Winning the future bid for $100 Access Points limits the intelligence and features you can place within -- before BOM hits $200+. By setting the price low, the perceived value is low. This year (2009) and 2010 will be challenging for the consumer Femto players. Their mission is to survive numerous trials and operators tinkering with the Femto business plan for consumers, while building strong relationships with major players who will ultimately be picked for the big roll-outs. Yes, that's how this market will roll.

Was it therefore a surprise when the same consumer players made a left turn and started to talk about Enterprise Femto? Of course not as these players need to find a profitable market segment where they can sell Femto APs at a higher price and therefore make up for the costly consumer business. However, the trap lies with the product strategy. The low-priced consumer APs cannot scale to the demands of the enterprise. You cannot simply name the consumer AP something else, paint it blue, and expect it to work within the enterprise.  A new board design is needed and advanced management and interference features demand a better product with tight integration with the gateway/aggregation router on or off premises. "Just" being a Femto AP vendor limits the control you have with operators and the enterprise alike. And, it gives you few cards to play with major OEMs. What's needed for the enterprise is: scalability, security and management features, QoS, advanced routing functions, local switching support and tight integration with the supporting RNC. In essence, the consumer Femto players need to design a radio network inside rather than a Wi-Fi-like access point.  The enterprise will demand that you include Wi-Fi.

This "left turn" requires a new approach and that requires more funding. The road to Femto success is very exciting but also riddled with "IAD"s (important access developments). It's great that the Femto market receives a lot of media attention. But, it takes more than hype to survive the economic situation of 2009-2010.

Today, ABI released its updated Femto vendor matrix (http://tinyurl.com/arsr2k). In light of RadioFrame's bold move to exit the AP business, the matrix could become an indicator of a race to the bottom. 

Show me the money, then I will believe in the Femto consumer business case.

Twitter/mobileinsider

Can You Hear What I Hear, Can You See What I See?

Fellow Mobile Insiders

This evening, after spending most of the day catching up with long-time friends who are industry and financial analysts in our industry (people who actually know what's going on inside the mobile industry), I went to my favorite web sites to catch up on news and developments.

Many of these sites reported (again) about Mobile WiMAX big plan for the future and that "next year" it will be everywhere (something we have heard since 2004). The latest, Clearwire will spend $1.5B to build out more cities in US next year. Really? We believe this?

To quote an article I read, "Clearwire brought in about US$20.5 million in revenue and lost $118 million"".... which means that another $1.5B will be spent to secure another 1M subs over the next two years while burning millions of more cash every quarter? When are investors going to start asking serious questions about the business plan or lack thereof?

I continue to be surprised by such big plans -- which are void of any plans for profitability.

As a nationwide network technology option, mobile WiMAX is paddling behind the big smoky steamship and coughing from lack of Oxygen behind the "MS HSPA-LTE."  

Today, Nokia reinforced its support for LTE. Their support for WiMAX was over last October. Their strong support of HSPA and soon, LTE devices, means that Samsung will be next to debark the sinking (mobile) WiMAX ship. Please do not misunderstand the pragmatic criticism. I do not dislike Mobile WiMAX. The technology had its chance in 2004 and 2005 but failed to deliver.

On the bright side, Clearwire has spectrum which is a valuable asset to some operators. VZ, T-Mobile and AT&T will need more spectrum by 2011 in time for LTE build-outs. So, I question any serious intent to spend any Capex whatsoever. It's a bluff which is why they are slow-rolling any build-outs. So, if you can hear what I hear and see what I see, then do what Google and others have done, write-down (off) mobile WiMax and move on.

twitter/mobileinsider

Mobile World Congress 2009 Viewpoint

With Mobile World Congress 2009 completed, my five net impressions of the show are as follows:

1. MWC 2009 was a solid event. There was much enthusiasm for the future. However, the sentiment is that next year’s show will be the testament of the industry’s economic viability and ability to weather a worldwide “financial Tsunami”

2.     The industry wants an Open Eco System for collaboration among vendors, standards bodies, and operators. Open API to allow for apps to traverse networks just like SMS. Any device, any network + common charger for devices, the focus for 2010 and beyond. What’s next, a common platform for charging our handheld devices? Oh, yes, that’s coming too (2012 target).

3. LTE infrastructure deployments may start as early as late this year with commercial services taking off in 2011-2012. Femto products and indoor wireless systems will play an important role to fill coverage and capacity holes for HSDPA/HSPA and support LTE in 2012 (WiMAX for Backhaul?)
4.  More Spectrum is needed but there will not likely be a uniform worldwide band for LTE by 2012 (400/700/900MHz, and 1.7 +2.6GHz the likely candidates? Time will tell)
5. No focus on Wi-Max…it’s now been relegated to backhaul and remote areas such as Africa. Today, MOTO also confirmed layoffs in its Wi-Max group which makes

Since Femto got some play at the conferenc, let me address the viability of this segment:

FEMTO and other indoor wireless solutions are indeed gaining momentum. This year will see 3 or maybe 4 major operators conduct controlled field trials (or labeled “commercial launch”) to test acceptance of these solutions to improve coverage and capacity for voice and broadband data and consumers and business’ willingness to pay. Unfortunately, the current startups will likely be pushed aside if the market takes off and hundreds of thousands of units are need. That’s when Cisco, Huawei, NSN, ZTE and Alcatel-Lucent step in.

Qualcomm is aiming to provide reference design in this space and given their 2H10 time-line, it may be well timed with market take off for this emerging segment of the industry.

Feedback from Telecom Italia was that consumers and business alike want an integrated ‘box’ for DSL/Wi-Fi/Femto. Netgear displayed an early demo box at the show. This “combo box” may be the home or SMB set-top box of the future (something Cisco will agree with)

This week, Brian Modoff of Deutsche Bank Securities sent out their Point of View on Femto.

“Most people we spoke with, including carriers, believe that the femto cell will not be a stand-alone box in the home. In many instances its features will be integrated into residential gateways or set-top boxes. This is particularly true of wireless operators who have wire-line assets.”

“We see this as toe-dipping by both carriers, seeking to fill some holes in their network coverage areas. The Samsung box can only handle voice and not data connections. It is also expensive, at an unsubsidized $250. Our understanding is that the box was built by Samsung’s handset unit, while femtocells under development by other vendors feature advanced network designs.”

“The biggest question around femtocells remains the operators’ interest.  Femtocell deployments still carry some thorny issues. The largest of these is how the femto will integrate back into the carriers’ core network. Much of this work has now been finished, with the standardization of the Iu-h interface, but final details are still being ironed out. Other technical issues, particularly interference management, are still being resolved.  If past roll-outs of new technologies are any good we would not be surprised if femtos get delayed by “unforeseen challenges” as we move through the year and into 2010.”

Overall, a good show in Barcelona. Next year, we hope to see less Thievery and no prostitutes on Las Rambla and more Police presence. More, cheaper, smarter mobile devices. A clear plan with regard to transitioning to a 20Mb mobile broadband experience around the world (and which bands to use). 

I’m sure, next year will be the year when personal video-casting makes its entry into handheld devices (early trials) which makes 2011 the year the entire mobile network crashed J

Twitter/mobileinsider

It's a Plane, It's a Toaster...It's a...Femto?

Michelle Donegan (Unstrung) is a brilliant journalist. She's insightful and she also is ON to the subtle shifts underway when it comes to vendors marketing gimmicks.

In her recent blog (http://tinyurl.com/ch3aoz ) she confesses that it is indeed confusing how the Femto and Pico Cell vendors are marketing and repositioning existing product lines -- or trying to create a differentiated product category. She brings up some simple and insightful questions as to what is what in Femto land.

Here's my crack at it (do not read too much into this)

1. Super femtocell: a beefed up picocell product repositioned to take advantage of Femto hype

2. Enterprise femtocell: anything a consumer vendor want it to be...at this point. Place an 'e' in front to existing Femto (Home Node B) and promise up to 8 users some time in the next 18 months and voila`, there's the enterprise femto (regardless of what the enterprise needs from a premise equipment).

3. Enterprise Femto becomes Picocell when a customer want a Pico cell solution and Femto when...you get the drift

4. When is Femto a small macro base stations?
A: It takes more than an AP, AP + Aggreagator to make an outside RAN. Some vendors will attempt to call a small Femto solution a Macro solution when LTE comes around. Scalability is key.

In the meantime, we have to witness Femto and Picocell vendors lobby for airtime and position products based on range, # of active users, time of day and time of month. It will only get worse before it gets better. Michelle - keep up the good reporting,

Twitter/mobileinsider

The Coming Mobile Network Capacity Problem

Today’s mobile operators are faced with a network capacity dilemma which will only become more critical as more smart phones and 3G-enabled laptops and devices are sold throughout the world.

The growing usage and penetration of high-speed mobile wireless services for both voice and data have created new challenges and opportunities for mobile operators, enterprises, and consumers. The requirement for always-on, high-speed communications both indoors and out is createing a capacity dilemma in the densely populated areas where enterprises are located (office parks) and high-rise office buildings in metropolitan areas. This capacity dilemma is not easily solved by adding more macro cell sites outside to ‘blast inside’ considering the expense and inefficiency of deployment in this way.

Attempts to provide the wireless coverage inside for voice and data has been cost prohibitive for mobile operators.  Wi-Fi, while offering a relatively low-cost wireless data access solution indoors, suffers from interference, integration issues, poor voice quality and issues with macro to AP and AP-to-AP handover. DAS and Pico-cell solutions are costly and require site and network planning. Existing solutions can either handle voice well or fixed data access– but not robust mobile voice and broadband inside the enterprise. UMA has its own issues and AP to AP handoff is not easily done over Wi-Fi (if at all). Show me a good working solution.

It will be interesting to see if FEMTO solutions can scale to handle dense areas without creating too much macro cell interference. Just adding a Femto AP does not solve capacity problems or coverage issues. Careful network planning is needed, otherwise we will face "Rogue" Femtos inside the enterprise (just like we experience with WiFi in '97 - '99 and beyond) and in dense areas which will only have a negative impact on the overal mobile network. The Femto Forum is on to the intereference issues and so are consumer femto vendors. The question is, when can they truely deliver a high capacity solution which address cost, coverage and capacity requirements for hundreds and thousands of people in a 1km by 1km square area?

SV Mobile Insider (twitter/mobileinsider

WiMAX Cloudy Future

In response to recent articles in Fortune (Feb 16 - Jon Fortt) and Forbes (Feb 6):

WiMAX: The Business Case Does Not Fly - anymore.

The promise of a $3B nationwide network is now more likely to cost $10B, something Clearwire does not have and Intel and Sprint would not underwrite (anymore). Current WiMAX users in the US should enjoy the spotty coverage where you have it, it's unlikely that we will see "NFL" city coverage any time soon. 

In 2003-4, Nextel envisioned a network costing $2.5-$3B with a mobile network from Flarion, a company acquired by Qualcomm in 2006. With the Sprint-Nextel merger, Intel promised to underwrite the WiMax technology and thus "WiMax everywhere" was what everyone envisioned by beginning of 2007. Then, with the issues and delays of mobile WiMAX, the vision became 2008... and here we are in 2009 and there are just a few cities with coverage. Clearwire (now includes Sprint/nextel WiMAX mobile broadband group), now realize that the "cheap" WiMAX is indeed as expensive as CDMA Do Rev A (or 3x more than FLASH-OFDM from 2005).

It is unlikely, given the economic climate, that Clearwire can deploy more cities without some serious funding or re-vector towards LTE and get economic incentives from the 3G community of vendors. But, the spectrum is indeed valuable. Maybe some major global operator will buy the spectrum in 2009-2010?

Regardless, WiMAX nationwide footprint is now a "pipe dream" and with HSDPA getting foothold with AT&T and CDMA/WCDMA operators looking to LTE commercial services in 2011-2012, the business case for mobile WiMAX has indeed failed.

MobileInsider (Twitter)