Tuesday, March 23, 2010

Future of wireless in India

India’s Mobile Revolution is still underway baffling even the most optimistic pundits who did not anticipate such unprecedented growth.  According to TRAI, India has more than 500 Million mobile subscribers now.  That number could be inflated because many old pre-paid connections may not be valid.   Even if we believe the number is 350 Million, it is still a very big market and it is growing strong adding 10-15 Million subscribers per month – bigger than entire mobile subscribers of some European countries.

India has delayed its 3G UMTS spectrum auctioning for many years now.  The question that I pose is whether India should even consider 3G UMTS or should we skip it altogether to move to FD-LTE?

UMTS or LTE

I recommend that India should hold off 3G UMTS spectrum auctioning for another year, completely skip 3G UMTS and instead embrace FD-LTE.

From prior experiences in deploying GSM networks, Indians think that old technology is always cheaper and better.  It is not always the case.   For example, a 20 GB memory in 1980 cost nearly USD 1 Million and was weighing 2000 Kg, whereas in 2010, it is costs USD 100 and weighs only 0.5 grams.  In such cases, no matter what you do, the old technology can never beat the new technology even with very high volumes.

We need to realize that in some cases, new technologies tend to be cheaper, advanced, efficient and reliable.  In this case, the older technology of 3G UMTS equipment is expensive, power hungry, bulky, less advanced and inefficient compared to newer and advanced FD-LTE equipment. 

 3G UMTS is based on WCDMA technology which is now able to give only up to 16 QAM (4 bits per symbol) where as FD-LTE based on OFDMA already delivers 64 QAM (6 bits per symbol) and soon will deliver 256 QAM (8 bits per symbol).   That means OFDMA clearly delivers more bits/Hz of spectrum.   Not only that, most modulations schemes lose bits in error correction schemes.  And here also OFDMA beats WCDMA.  OFDMA based technologies incorporate other advanced techniques like MIMO, Smart Antenna, sub-channelization, etc, which increase the efficiency of these modulation in delivering high throughputs compared to WCDMA.   Clearly WCDMA is less advanced and less efficient compared to OFDMA technologies.

Moreover, WCDMA involves chip-rate processing where every symbol will have to be multiplied by a large number, as large as 3,840,000 chips, requiring huge processing hardware, make the equipment and handsets power hungry that means your battery drains faster, makes them bulkier, and expensive.  Also, the base Stations for current WCDMA sites use up to 6000 W of power, whereas one can build FD-LTE base stations that use 500 W or less.  In many emerging countries this 10X advantage in power consumption makes a huge difference because most cell sites take up huge cost in a/c housing, generators, battery backup, power management systems, etc.

Indians should understand that older technology is not always better.  India has to seriously look at advantages of FD-LTE and adopt it completely skipping 3G UMTS.   With this move and other incentives discussed below, India can also spawn a local ecosystem of companies that can launch global telecom equipment companies.  It can use this opportunity to create Huawei and ZTE of India.

Separate BWA from 3G

While India is rapidly adding mobile subscribers, it has lagged behind in broadband penetration, mostly because it was still a wired deployment.  With advent of wireless access, the current 6.25 Million broadband subscriber base can go up to 30 to 40 Million in the next 4 years bringing in the Broadband Revolution in India.   Currently, the broadband penetration in India has been floundering because of lack of comprehensive and independent policy on Broadband Wireless Access (BWA) spectrum. 

I strongly recommend BWA spectrum policy should be separated from 3G UMTS spectrum policy, and the BWA spectrum should be auctioned independently.  In addition, India should clear up and allocate at least another 250 MHz of spectrum between 5.1-5.8 GHz range from the current 50 MHz and mark it as unlicensed.  This will spawn many entrepreneurs, many tier-2, tier-3, and tier-4 ISPs who will start giving broadband to homes, offices and enterprises without going through spectrum purchases.  This will result in unprecedented broadband activity in India. 

Currently, the 3.3-3.4 GHz spectrum is given to Indian operators on a trial basis.  Unless India spells its strategy clear it does not allow operators to take big bets because they do not know if the spectrum will be taken away from them anytime.  Meanwhile, India should add another 100 MHz from 3.4 – 3.5 GHz for BWA.

With additional 80 MHz spectrum in 2.3-2.5 GHz earmarked for BWA, India will have three ranges of spectrum, in 2.x, 3.x, and 5.x ranges making it possible for tier-1 to tier-4 ISPs to play various roles, some targeting urban market, some rural market, some enterprise market, some residential market.  This will immediately add 30 to 40 Million subscribers in the next 4 years, resulting in USD 5 to 6 Billion equipment installation.  

China like incentives to local companies

India should give incentives to local companies to play a dominant role in the domestic market.  Otherwise, the bulk of telecom equipment revenue is now going out of India.  The total telecom equipment market is USD 24.7 Billion in 2008-09 growing 20% from the previous year.  Most of this market is captured by foreign companies.  For example, cellular wireless infrastructure comprising 30% of the total telecom equipment market is taken up by international giants.

Imagine if domestic telecom equipment industry increases its share even by 20% of this total telecom equipment market.  That adds nearly USD 5 Billion market immediately.   And yet, there is no incentive to Indian companies to do that.  The only quota system we had was reserved for ITI, which sold Alcatel-Lucent equipment depriving Indian companies further.  Such flawed systems combined with low-risk capital investors have not enabled Indian telecom companies. 

India should not neglect active infrastructure

Most Indian analysts and industry experts tend to believe that mobile handset market is more attractive compared to active infrastructure because of the volumes.  However, the actual numbers tell a different story.   Carrier Equipment comprised a whopping 64% of the total Telecom equipment market in 2008-09 while the Mobile Handset comprised only 23%.  Not only that, the share of Carrier Equipment has been steadily increasing from 56% in 2006-07, 60% in 2007-08, to 64% in 2008-09, whereas the share of Mobile handsets is decreasing, from 31% in 2006-07, 26% in 2007-08, to 23% in 2008-09.

Most telecom companies in India continue to concentrate on passive and paraphernalia infrastructure which is only 5% of the total telecom equipment.  Apart from exceptions like Tejas Networks raking in some portion of optical network equipment, we don’t see many Indians companies playing a role in active infrastructure.  Companies like Huawei, ZTE, Nokia Siemens, Ericsson, etc, take up bulk of the revenues coming from this market where no Indian company has any presence. 

Even if we think that the telecom market in India grows at a nominal 15% per year, the opportunity over the next 5 years is USD 200 Billion.   How much of it is going to come to India?  Unless we do something drastic and dramatic, bulk of it will taken up by foreign companies. 

I strongly recommend that India should provide incentives, subsidies, quotas and loans to domestic companies focusing on active telecom infrastructure.  In absence of investments from private equity firms, Indian government should set up fund to invest in domestic telecom equipment companies so that they take on this challenge to capture substantial market share in India.

Once these telecom companies succeed in Indian domestic market, they can play a dominant role in the world markets, like Huawei is doing right now.  They will also cater to Indian defense communication equipment which in itself is a very big market.  


Terms Explained:

TRAI – Telecom Regulatory Authority India
3G UMTS – 3rd Generation Universal Mobile Telecommunication System, an extension to 2G GSM family (GSM->GPRS->EDGE->UMTS)
LTE – 4th Generation (?) Long Term Evolution, further extension of GSM family (GSM->GPRS->EDGE->UMTS->HSPA->LTE)
WCDMA – Wideband Code Division Multiple Access (underlying modulation technique used in 3G UMTS and CDMA 2000)
OFDMA – Orthogonal Frequency Division Multiple Access (underlying modulation technique used in LTE and Mobile WiMAX)
QAM – Quadrature Amplitude Modulation, higher the modulation scheme, more throughput.
 



No comments: