Tuesday, June 10, 2008

Ten Challenges for New Born 3G iPhone

Ten Challenges for New Born 3G iPhone

The US economy is in recession, and it will take 5~8 years before we have another economy booming. It is a tough time for any technology movement. The gas price will go up to $7 per gallon by the end of this year, and the US dollar keeps devaluating. This is definitely not as good time as that when iPod came to the world. In addition, the following ten challenges will stay hard in the way of 3G iPhone movement:

1. The service cost is the most critical issue for the mobile users. While Jobs promoted the 3G speed as similar to the WiFi (which is FREE in many cities), he forgot an important issue for the not-rich people, the cost of 3G service! I do not believe AT&T can offer a cheap package for this 3G service, and in fact, most American people can not afford pay the expensive 3G services. While almost everything becomes expensive, people need to cut budget to save dollars for more milk, gas and food, the 3G is most probably in the budget cut list because it is not a must.

2. More iPhone technologies are uncovered which makes copy much easy. In Taiwan and Shenzhen, you can pay $5000 to buy a complete hardware and software package for iPhone systems. Though may not be updated with newest version, it is enough for a 15 years' senior engineer to redo the system again. Touch screen technology, Visual rotating technology and Push technology, claimed as Apple's core technologies, now become a public know-how in many countries. Patents are only applied to gentlemen, not to copyguy especially there is no enforcement in many countries.

3. More and more iPhones will be unlocked across the global. Though Jobs claims more secured technology in locking the phone, but as long as it is locked by engineer, another engineer will be able to unlock it eventually which is the EE law. I witnessed some Ph.Ds and engineers were very smart to be able to reverse engineering of a very complex 128K ATM Swicthing Matrix within a month, so whatelse can not be decoded. The reverse engineering market in Asia is a tens of billion dollars' business, and over ten million professionals are working on this industry every single day. In China, thousands of Ph.D candidates are working on analyzing iPhone as their Ph.D thesis, and their results are amazing.

4. Apple is not a mobile wireless company, and most iPhone technologies are borrowed or leased. iPhone is different from iPod as it is a phone. The mobile industry is already full of fires and wars. Without so much wireless technologies and influence, how can Apple survive in long term strategy is a big question. 3G was a ten year's ago old technology, and had been struggling for long time, and I do not believe it is going to success (like ISDN in old days). People are working on LTE (long term evolution) and possible 4G on open wireless architecture (OWA) of wireless convergence, but Apple just started 3G which is really funny. The AT&T so-called 3G, who knows how many peole really uses it and how long it can survive, is a very risky business as well. In case this so-called 3G is successful, the next question is how to deal with the handover between WiFi and 3G, how to avoid AT&T forcing users to use 3G networks even in the free WiFi networks because AT&T wants to charge users more to its 3G networks. The iPhone is locked by AT&T, and therefore AT&T determines users' payments based on AT&T's interests, and everyone is hostaged. In case all these issues are solved, then Apple will be sued by lots of companies with different patents in wireless fields, and Apple do not have so many IPs to fight back. Remembered before 1996, Qualcomm did not have many lawsuits, but after its CDMA products started to make much money, lawsuits became falling snows. Nevertherless, Qualcomm is an wireless technology leader. But Apple does not have wireless IPs, most relying on Interdigital and its partners. Personally, I listed around 80 patents which are threat to iPhone's 3G business. Let's see!

5. 3G iPhone is not going to make much money. In my 09/06/2007 Blog, I told everyone that 8G iPhone's BOM cost is $137.62, so I predicted the price should be less than $199. Considering large volume and high-cost of 3G version, the 3G iPhone's BOM cost is around $149. So Apple does not make much money from the 3G iPhone products. Of course, with the volume increases, the BOM can be low as $100, but takes a while. Meanwhile, the fake iPhone will continue to reduce price which generates high pressure to iPhone sale price. Further, please remember, 3G iPhone RF provider - Infineon has a project with Huawei, etc to deliver a very cheap UMTS chipsets which may help fake iPhone (or called other names) to be very cheap. Also, many samll 3G companies such as Sandbridge, etc may offer much cheap chipsets and make "iPhone" even more cheap in "Water" markets.

6. More fake iPhones than legal iPhones soon. With the increasingly popularity of iPhone, more and more peoples are investing and developing fake iPhones. Without marketing costs (thanks to Apple for global marketing), these small companies are easy to survive at the moment. The iPhone hardware and software are available in many countries, and the key chip vendors are open for anyone, so it is easy for copy and assembling. Though some functions do not work well, but many people may never use it in life time. Because it is cheap and not locked, there are always markets. Currently there are over one million fake iPhones in the market.

7. Apple needs more R&D in wireless. iPhone is so polular, but Apple never has a R&D in wireless. If you search in the IEEE Xplore, there is no wireless/mobile paper from Apple. Relying on partnership is just a short-time strategy only. For long-term success, you need your own R&D. I never heard a company can survive long time without its own R&D. Qualcomm, Nokia, Samsung, etc have tens of thousands of R&D papers every year, and Apple needs to think about it seriously. When Jobs is enjoying its 3G baby now, the world almost left 3G era to step into the 4G era, no matter how to define 4G, it is a big movement towards future.

8. 3G iPhone's architecture needs big improvement. While 3G iPhone has lots of applications much beyond a iPod, it forgets it is a wireless and mobile phone. The transmission and spectrum efficieny is mission critical for a multimedia mobilephone. The current design architecture does not allow the phone to transmit data and video so effective over the air, and the transceiver was originally not designed for iPod-like traffics. When the capacity increases, all the problem comes out.

9. 3G iPhone network optimization will take long time. Qualcomm took ten years to increase the CDMA capacity from 20% to 60%. Though AT&T started 3G deployment years ago, but it takes long time to increase the capacity. When the capacity is low, everything becomes easy and high-speed. When the capacity increases, lots of problems with mobility and broadband transmission will come out, and especially CDMA is not optimized for high-speed tranmission. It will challenge the user experience with iPhone. Anway, it is a wireless phone with seamless mobility.

10. 3G iPhone is not so secured. The current 3G version hardware and software architecure do not have good security measures with implementation in the MAC and PHY layer, meaning the system is still exposed for attack, though there are some developments in the high-layers. In wireline networks, high-layer protection is enough because the transmission medium (cable, fibre optic) is stable and not expose to the public. But in commercial mobile communications, unless secured means in MAC and PHY layers, everything transmitted is open to the others, and "experts" can easily park their cars outside of your house, and get hold of your tranmitted data to their equipments for further analysis. VPN in this 3G networks is not so secured, unless improvement in MAC/PHY is done. Unfortunately, the Infienon S-Gold chipstes and other suppotrting chipstes do not support such security features in this 3G iPhone version.
Anyway, we need more work on this new "old" baby, and let's help it to move the business forward.


Prof. Willie W. LU
Former Senior Principal Wireless Architect
Infineon Technologies AG
Wireless Solution Provider of 3G iPhone

Hardware Disclosure on iPhone 3G Device

Hardware Disclosure on iPhone 3G Device

Provided by USCWC, China 4G R&D Centerfor more information on detailed Hardware and Software architecture, please contact nancy@cwc.us.
iPhone 3G Version:

* Infineon PMB6952 / S-GOLD3 6-Freqs UMTS / HSDPA Transceiver
* Murata LMRX3JCA-479 triband amplifier
* Sony SP9T GSM / UMTS mode switch
* ARM 1176JZF-S main processor
* Skyworks 77427 - UMTS / HSDPA TX 1900MHz, RX 2100MHz
* Skyworks 77414 - UMTS / HSDPA 1900MHz
* Skyworks 77413 - UMTS / HSDPA 850MHz
* Layout: n82ap (vs. m68ap in version 1)
* UMTS power-saver option
* A-GPS module

To order the software architecture and documents, please contact nancy@cwc.us.

Wednesday, March 26, 2008

Story Behind the 700MHz Spectrum Action

Story Behind the 700MHz Spectrum Action

700MHz spectrum itself is not so important as you image. But the story behind the 700M is much more important and critical.

700M spectrum is good for long distance radio transmission, but not appropriate for broadband highspeed transmission, and therefore it is mostly for voice and message services for cellular mobile communications. If you use 700M for fixed or local wireless access, the datarate can be high and up to tens of Mps, but with very limited mobility. Therefore, from the technical point of view, 700M is no different from 800M/900M which we have used in cellular mobile for over twenty years.

700M is also not the last spectrum to auction. We have more spectrum on the way including new ITU IMT-Adv band and future 4G spectrum, most probably around 2.5~3.5 bands.

The most important issue behind this 700M case is the attemping to open up the US mobile communication markets, which has been strongly urged by the governmant and the general public.

As you may know, US mobile markets are extremely closed. Users pay the handphone or free phone with long-term contracts, but their phones are still locked by the operators. A 12-months or 24-months service-contract basically PAY the mobile phone which is not FREE though the users did not pay it when signing-up. American people have been fooled by the telecom operators for long time. If you buy the TV and it is locked to Comcast only, do you accept this situation?

In 1997, Chinese law said "locking a mobile phone without subscriber's permission is a national crime". My US mobile phone has been locked since 2002, and is permanently locked now. I called the operator for more than 10 times to request unlock,was never successful. Eventually a customer service representative called me back to tell me follow her instruction to unlock the phone - she told me input 3-digits security code to unlock. After I tried twice with failure, she told me sorry that she gave the wrong code. After I tried the third time still not successful, she said sorry again and need to check with the technical support team. After 30 minutes, she called me back and said sorry that my phone had been permanently locked. I was very angry and filed complains to the service provider and operator (you can guess this operator). I visited the local store of the operator, and complained with the manager. He told me it was nothing to do with the operator because of three: first, I did input the wrong code myselfin unlocking the phone and it was my own responsibility, and the customer representative just helped me without taking any responsibility. Secondly, the phone has been permanently locked by the phone vendor (in this case, Sony-Ericsson), I should contact the phone vendor instead of the operator. Third, if I needed to file legal complains, I should talk to operator's lawyers rather than the local store. He further added, the operator has hundreds of lawyers, and he welcome any lawsuits.

Another story behind the 700MHz is the battle between the telecom industry and the computer industry, basically between the east coast and the west coast industries. As the future mobile device will be converging the computing, networking with wireless mobility, it is a very good opportunity for the computer industries to enter the wireless markets, especially to extend the Internet business to the mobile wireless Internet markets. However, the battle will last a while if the government does not stand out to enforce either side because the traditional telecom industry controls the infrastructure, and the netwok & computer industry focus on the user or access equipment segments. To be honest, this is a governmental regulation issue rather than a technical issue. Both West Europe, Japan, China and many other countries solved this issue very well, why can't America?

Last, but not least, this 700MHz will start a new initiative of dynamic spectrum planning in response to the continued exhausting of available spectrum. As no single wireless standard can do both broadband high-speed and seamless mobility, multiple wireless standards are to be converged and integrated into common mobile device in future which requires the spectrum be allocated dynamically and openly.

The global mobile markets are pretty open now, but US is still one of the very few countries locking the users' mobile phone. Openning up our mobile communications is a trend, not an option, and it is just a matter of time.

Prof. Willie W. Lu
Director, USCWC
Palo Alto, CA

Sunday, March 23, 2008

Sprint WiMax-Xohm Will Fail Again

Sprint WiMax-Xohm Will Fail Again

by Tao Zhang, Reporter, World Mobile Congress

Sprint had been trying to kickoff broadband wireless business for over decade. Starting from wireless and mobile ATM, to MMDS and LMDS, Sprint failed too many times. Sprint would repeat the failure story again in this WiMax trial for reasons as follows:

1. Sprint's management is too questionable to manage such a big technology movement;
2. Sprint does not have enough capitals to deploy such a money-burning project;
3. No big partners including Intel will invest dollars in Sprint Xohm movement which places Spring in a very risky situation;
4. There are still lots of technical arguments about Sprint Xohm's seamless mobility features because no single wireless standard can do both broadband high-speed and seamless mobility. Therefore, Xohm may become another lesson of MMDS if it only supports fixed broadband wireless access or limited-mobility wireless access;
5. The business model of Xohm is not clear in terms of how open the Xohm networks are and what is the targeted market segment - mobile handphone market vs. wireless laptop market?
6. The international standards are not clear for Xohm solution - LTE is gaining worldwide against WiMax, and some countries including China may support LTE instead of WiMax;
7. The long-term strategy of Xohm movement is not clear in terms of convergence of multiple wireless standards and open wireless infrastructure.

All these issues make Sprint Xohm very risky and challenging for investors and users. Therefore, I believe Sprint feels hard to deploy this network successfully. Without huge investment from key partners which is even harder at this economic situation, Sprint will fail in this business again.

WiMax is a very good technology, but Sprint is just not qualified to deply it.

Tao Zhang, Reporter
World Mobile Congress

Friday, September 28, 2007

What's Going On for WiMax?

What's Going On for WiMax?

I have been asked the same question in the last 2 weeks by over 200 investors, attorneys, analysts and plicy-makers because they trust my views are very different from the public publication or public release.

Yes, I am an academic people plus was a low level technical labor as Chief wireless architect of Infineon Technologies. I started broadband wireless access (BWA) technology since 1992 and owned couple of Wireless mobile ATM patents, and later started IEEE802.16 together with Roger, Jim, Brian, etc in march 1999.

Before we discuss on any issues of Wimax, let's sit down and calm down for a while - forget any political, marketing, and media hypos, and think it over carefully as a student, rather than as a businessman or salesman.

WiMax is not a new baby! WiMax is already at least 15 years'old teenage. We all know Teenage is a very critical time period in life, and we need enough education to help set his/her long-term development strategy.

WiMax is just a new name for IEEE802.16 standard, one of the standards for Broadband Wireless Access (BWA) technologies. In history, BWA included Wireless ATM (autonomous transfer mode), wireless mobile ATM, HyperAccess, LMDS, MMDS - some are IEEE, and some are non-IEEE standards.

IEEE802.16 came from an idea of Cable Modern (DOCSIS based) in later 1998, and 802.16.1 on LMDS and 802.16.3 on MMDS in early 1999 when we started the 802.16 standardization activities in Boulder, CO, the virgin home of WiMax. Later 802.16.1/3 changed to 802.16.a...

Before we discuss about the WiMax, we need to analyze why the previous Wireless Mobile ATM and LMDS/MMDS were totally dead. The first issue I believe is the COST and PRICE. Do you really can provide a low cost (I should not say cheap) solution to the users, and how cheap the service is?

Some people focus on the under-developed emerging markets such as China rural areas or India rural areas where the telephone line is expensive to install. But people still use the American logics to think about the cost in China. In US, the telephone line installation is expensive because the labor and construction are expensive. In US, if you want to install a 5-miles phone line to the rural area house, it may cost $1000 per line. But in China rural area such as Gansu, Jiangxi, Xi'an, etc, we only cost $20 per line to install a 5-miles phone line to the rural home because the labor and construction is extremely cheap, and we do not need a permission from the household to install a phone line in the property or set up an antenna on top of the building (in rural areas). If these areas can not afford to pay $20 per line to have phone line, do you think they can aford to buy the WiMax equipment?

But we do have huge markets for WiMax in China - the Governement markets and Community markets. China gov't controls almost 85% of the ICT industries and there are many opportunities for WiMax systems such as Traffic department, transportation, customs, business parks, oil plants, banks, ATM stations, schools, city halls, etc whci are all gov't markets. The issue is: How much cheap you can provide for WiMAX solution? As long as the cost is low, people will love it!

Personally, I believe Intel is very smart and will win in this race. Do you think Intel is a WiMax company? If you think so, you are stupid!

Intel is a IC company with endless innovations and inventions in key technology of IC. Intel will leap ahead, but not heading to WiMax! Intel is absolutely targetting open architecture, and have WiMax as a testing example!

Remebered the PABX race in early 90s, only one company made huge money - that was Mitel shipping the TDM swicth chip 24/7 to every corners of the world.

WIMAX is a very good technology, especially in fixed and nomadic WiMAX solutions. But for seamless Mobile WiMax, we still have lots of technical issues to be solved, and are very difficult to be solved. I always believe there is NO single wireless standard that can do everything both broadband and seamless mobile. We need different wireless standards to complement each other for the converged solutions.

For general mobile Wimax, the term "mobile" is very confusing and misleading. What do you mean by "mobile"? There are several definitions of "mobile":1. Portable, such as mobile computing.2. Moveable, such as mobile house.3. Free mobile (or called seamless mobile), such as mobile phone.

So in the current IEEE802.16e, the mobile WiMax is still limited to "portable" or very limited mobile.

Why I said seamless (free) mobile WImax is very difficult or even impossible? When the frequency goes high, the cell size gets very small, and the transmission is limited to line-of-sight. The frequent handovers between various base-stations become very difficult especially when the network capacity increases. There is no well accepted solution yet to support the free mobile handover control protocols in broadband wireless IP networks. The traditional GSM/TDMA and CDMA protocols do not support such high-frequent and dense handovers in such small cells with non-circuit-switch (NCS) connections.

But on the other side, mobile may just become one feature for the WiMax, just like camera feature in a mobile phone - it is available whenever you want to use it, but mostly you never use it unless it is too cheap or free cost.

I fully understand that people want to distinguish mobile WiMax with previous MMDS, etc, otherwise it may repeat the same lessions of MMDS except the cost and performance improve a lot.

Last, any new mobile standard needs a lucky place (Fengshui in Chinese) and lucky time to save it at the begining. GSM was saved by God (God Saved Mobile), CDMA was saved by Korea, and who is going to save mobile WiMax?

Anyway, WiMax is a very good technology, I believe.

To be continued .... as we are landing to the airport.

by Willie W. Lu
in Flight

Thursday, September 6, 2007

How much can Apple low down the iPhone price

How much can Apple low down the iPhone price?

As I said in July in the Silicon Valley's Churchill Club, and also repeated in my Blog on Aug. 6th, Steve Jobs has no other choice but to cut its iPhone price because it is TOO expensive, and its competitors offered two low price for similar phone. In China and Taiwan, several iPhone-like mobile phone only costs $150 or even $100, though not so shinny, but almost all functions work very well.

I visited an iPhone "development" company in Shanghai, and was told that the 8G iPhone BOM cost is only $137.62 which is surprisingly low. I am not sure whether that is a Chinese version iPhone or US model iPhone, but the fact is, it is an iPhone and works exactly same as the one sold in AT&T, plus everything is unlocked.

iPhone is a symbol of fashion and power, and everything (in the industry) is talking about this angel across the global. The 8G iPhone was $599, but now is $399. Do you think it is a reasonable price?

If the BOM cost is $137.62, the total phone cost should be less than $160. hence, the $399 may still carry 150% profit margin. I am a mobile phone architect for over decade, the GSM phone BOM cost can be low to $13, and the CDMA phone can be at $15 in China. The Chinese designed "iPod" (sold in Shenzhen and Beijing) BOM can be low at $15 as well. So it is really subject to what kind of "iPhone" you are talking about.

After the iPhone hardware and software architectures are uncovered, the phone is very normal except the display components and the application processing modules are quite aggressive. But on the wireless phone side, the BOM is very cheap. Some Taiwanese companies are redesigning the expensive components of the iPhone in order to further low down the iPhone cost to less than $80 in BOM, and less than $100 in total.

When the iPhone goes international, and based on Jobs's strategy, Apple will sell the iPhone to Asia in 2008, how to deal with the low-priced local "iPhone" in many Asian countries becomes the very tough challenge for Apple. As you all know that, IPR enforcement is extremely difficult in some countries, how can you protect your phone from being copied, reverse-engineered, assembled? There are hundreds of Ph.Ds working on iPhone right now, and these people are very smart - believe me!

Personally, I expect Jobs will continue to low down the iPhone price, probably in 4~6 months time frame.

Tuesday, September 4, 2007

iPhone on the way to Open Wireless Architecture (OWA)

iPhone on the way to Open Wireless Architecture (OWA)

When the New York teenage unlcoked the iPhone, it sent the message to the world: iPhone is one the way to liberate itself! Basically, Apple never cares to lock or unlock the phone, as it makes profits by selling the phone, not selling the services. Furthermore, Steve Jobs hopes it is fully unlocked so that anyone can easily use it anywhere and anytime, same experience as iPod.

Apple computer is in Cupertino, the heart of Silicon Valley and also my hometown, where WiFi is a free service across the city of Cupertino, and the speed is quite high. There is hugh demand to have iPhone working in WiFi networks which is totally unlocked and free service. From Cupertino to Palo Alto, most Furtune 500 IT executives and big guys are living there, and they push Steve Jobs to consider unlocked iPhones for WiFi and other broadband wireless access networks.

Furthermore, Cisco, Google and Intel also pushing hard the upcoming mobile phone business supporting unlocked and multiple wireless standards to offer the truly service-oriented information delivery platform.

Therefore, the game is pretty clear for Apple: iPhone needs to go open wireless architecture (OWA) and it is the best time to do so:

1. iPhone was lunched on AT&T GSM networks with iPhone locked,
2. iPhone starts to operate in WiFi networks and can be unlocked as requested by users,
3. iPhone will be operative in 3G networks and can be unlocked as requested by users,
4. iPhone will be operative in WiMax networks and can be unlocked as requested by users,
5. iPhone will support OWA Phase I with 2-in-1 or 3-in-1 air interfaces in the same iPhone,
6. iPhone will support OWA Phase II with open air interface standards in the SIM Card or AI (Air-Interfaces) Card.

Once iPhone is going for OWA openness, it is ready for VoIP, mobile searching and full landscape of mobile applications and services. In long term, Apple, Cisco, Google, Oracle and Intel, etc will all work together to create the truly service-oriented 4G (fourth generation) mobile communications.

Keep your eyes open for iPhone as it goes for open wireless architecture in long run!