Broadband has long been like the proverbial Godot for Indian Internet users–it
is supposed to be there but it never comes. As users–both business and
residential–began yearning for it, they were often promised of it, but a fat
pipe always remained, and still remains, elusive. Paradoxically, this is the
state of affairs in a country where almost all the broadband technologies–be
it DSL, cable modem, ISDN or Ethernet–have been deployed commercially. While
some of the broadband services failed because of flawed business models (like
high subscription charges and high CPE cost), services like DSL did not take off
because of regulatory hurdles, while others like cable modem failed because the
technology wasn’t suitable. And in all these, non-existence of a critical mass
of users has certainly been a major issue.
Interestingly,
despite not so friendly regulatory regime, mostly private players are currently
offering whatever broadband service exists in India. However, none of them have
made any significant impact. Significantly, incumbents BSNL and MTNL never
really thought of deploying broadband services. This lack of interest of
incumbents has surely contributed to this long Indian wait for broadband. It was
only recently that BSNL took an initiative. However, it is still unclear what
strategy it is looking at.
Is DSL the Way to Broadband?
A number of broadband technologies have long been there. However, none has
been as successful as DSL. In broadband markets across the world, DSL has been
fast emerging as the residential broadband technology of choice. In fact, DSL
has been regularly eating into the cable modem broadband market in countries
like US. However, in India DSL has remained a non-starter, largely because
incumbents who own millions of the copper loops have never really taken it
seriously. In the context of the global success of DSL, BSNL or for that matter
MTNL or any of the new private fixed service providers could do well by
considering DSL for taking broadband to homes. That apart, the very nature of
the DSL technology and the falling cost of DSL equipment makes it the best
option for broadband-hungry but price-sensitive Indian users.
|
Falling Equipment Costs
Today, it would not cost more than $100—150 including the cost of CPE for
taking DSL to a residential subscriber. The same used to cost around $250—300
only a year ago. This clearly shows a downward trend in the DSL equipment
prices. In fact, an important factor in the growth of DSL in the Asia-Pacific
region has been the fierce competition among the DSL equipment manufacturers
based in China and Taiwan. That has pushed the per-port price of DSLAM from some
of the vendors to $82. Leading global vendors like Alcatel, Siemens, Lucent and
Cisco have been under increasing pressure from equipment vendors in China,
Korea, Japan and Taiwan. "After including duties of around 30 percent, DSL
equipment should cost in the range of $125 per line in India," Ruchir
Godura of UTStarcom observes.
|
Prices of ADSL modems too have been falling steadily, so much so that they
are becoming commoditized products. This is forcing manufacturers to make money
on volumes rather than on value-adds. ADSL modems are now available in Taiwan
for $50—60 and are expected to become cheaper in the coming year as ADSL
becomes even more widespread in the region and competition among vendors becomes
fiercer. In fact, in many ways most of the US and European vendors of DSL
equipment have been pushed out of the Asia-Pacific market because they have not
been able to match the prices offered by the regional manufacturers. The input
cost of manufacturing has been so competitive in the region that China has
emerged as one of the most important manufacturing base for Alcatel, the largest
DSL equipment vendor globally. "Our company Sanghai Bell has been able to
manufacture DSL equipment at Asian prices," points out Alcatel’s Bernard
Grave.
Some of the vendors are offering ADSL modems with built-in routers that have
VPN and voice capabilities. DSL CPE manufacturers are also integrating wireless
technologies like 801.11 to accommodate in-premises data distribution and
Internet sharing on existing media types such as wireless, phone line, and power
line. The next evolution in home networking will address the serious problem of
in-premises multimedia distribution using existing media types, and will require
a solution that delivers guaranteed QoS. All this has transformed DSL modems
into integrated access devices (IAD). However, these additional functionalities
are not required for an average user, at least here in India at this juncture.
DSL Success Stories From Around the World |
Deployment cost and suitability of technology apart, a combination of other factors that include government initiatives like promoting use of broadband for education (like in South Korea), deregulation (primarily unbundling of the local loop and allowing private sector participation), demographics (high PC and internet population), low tariffs, services and vigorous competition has made DSL as the residential broadband technology of choice across the world. Currently, there are around 28.5 million DSL subscribers which amounts to around 37 per cent of global broadband connections. The Asia Pacific region (primarily South Korea, Japan. Taiwan and China) account for almost half (47 per cent) of the total DSL subscribers, while Europe has 29 per cent and the US 21 per cent. DSL deployment has helped broadband Internet grow tremendously in South Korea, Japan and China (in that order) in the past one year. The growth has been particularly exceptional in South Korea where between December 2000 and September 2002 the number of Internet subscribers via ADSL increased from 2.35 million to over 5 million, representing an increase of around 113 per cent. Japan has around 4 million DSL subscribers. China currently has 1 million DSL subscribers. DSL growth has also been encouraging in Western Europe where incumbent service providers are focussed on DSL to offset declining voice revenues and a combination of factors have pushed DSL on the forefront of the broadband push. In Central and Eastern Europe too, DSL is likely to emerge as the dominant broadband technology in the coming years. In 2001, xDSL surpassed cable modem as the predominant standard for broadband in Latin America. By 2007, there will be more broadband subscribers using xDSL in the region than all other access technologies combined. Similar is the case in the United States where DSL’s share in the broadband services market is growing fast. |
The Technology
The falling cost of equipment apart, the very nature of the DSL technology
has also been the key to the growing popularity of DSL among service providers
across the world. Even though DSL utilizes standard telephone lines, users can
move data along those lines at higher speeds. And this is reason why it is
particularly suitable for incumbent fixed line telcos like BSNL and MTNL who own
millions of last mile copper loops. The service works best for users who live
within three-and-a-half miles from the telephone company’s central office or
where the DSL equipment is located – the more the subscriber is closer to the
central office the better speed he gets. And even if the subscriber were farther
then the prescribed distance from the central office, he would get a far better
speed than a dial up or a cable modem. This is because DSL is rate adaptive,
that is, it adapts the communication speed to the user’s modem and distance
from the service provider’s point-of-presence. In fact, Rate Adaptive DSL is
being touted as the best bet for residential broadband as it provides maximum
bidirectional bandwidth from a DSL circuit.
Comparison of DSL and other technologies |
||||
Parameters | ADSL | Dial Up |
Cable Modem |
ISDN |
Speed | ADSL offers dedicated speeds (up to 6 Mbps downstream and up to 640 Kbps upstream) |
Maximum speed only 56 Kbps |
Cable modem exists on shared network, therefore speed performance unpredictable, speed dependent on network traffic volume |
Offers guaranteed speeds only up to 128 Kbps (running on two channels at 64 Kbps each) |
Flexibility | Fully scalable with wide range of potential speeds |
Bandwidth limitation of 56 Kbps makes upgrade impossible |
– | Not scalable as maximum bandwidth is 128 Kbps |
Reliability | Since ADSL offers dedicated access, it avoids the time consuming process of dial in for the Internet. Low risk of ownership due to minimal hardware requirement. |
Dial up access is often faced with tedious process of dialling in for Internet access. |
Shared access makes speed performance unreliable and unpredictable |
ISDN has a high risk of ownership due to extensive hardware requirement |
Price | DSL is priced on a flat monthly rate with no additional usage for unlimited access/telephone charges. Also, one connection can be cost-effectively shared by multiple users |
Dial up access can have monthly telephone charges |
– | ISDN can have costly monthly phone usage charge in addition to Internet access |
Accessibility | As ADSL rides on existing copper phone lines, it can reach 100 per cent of the telephone users |
Widespread Access |
Cable modem utilizes young network infrastructure that is shown to have sporadic and inconsistent service availability. |
– |
One of the most notable features of DSL service is that subscribers get a
single secure line that directly connects to the service provider and users get
dedicated bandwidth as the line is not shared with any other user. All DSL
technologies exploit the available spectrum in copper telephone lines, employing
advanced modulation techniques to carry high-speed data transmissions. The basic
differences among the DSL technologies lie in their communication speed,
operating distance, and suitable applications. Apart from using the existing
copper lines, all DSL technologies offer certain other benefits like
partitioning voice and data traffic. And since the services can be rolled out on
the existing infrastructure and requires only minimal changes in the access
facilities, service rollouts are not risky bets for service providers. Moreover,
service providers can easily adapt their existing customer support, billing, and
installation activities to support DSL services.
|
All DSL technologies require a similar configuration of user and service
provider equipment. At the user end, DSL access requires a copper phone line
that is connected to a DSL modem (or router). At the service provider’s
central office or PoP, DSL access multiplexer (DSLAM) connects multiple users
and passes the transmissions to their respective voice and data networks. DSLAM
also separates incoming voice and data signals and directs them onto appropriate
carrier networks.
Of the various DSL technologies a la IDSL, SDSL, HDSL, ADSL and VDSL, ADSL
has emerged as the technology that is driving residential broadband. The nature
of ADSL technology makes it fit for Web browsing and other Internet applications
in which users download far more information than they upload. It transmits a
much higher downstream rate than upstream.
This asymmetric nature of ADSL also means that it is not optimized to support
business applications that require roughly equal amount of bandwidth upstream
and downstream. ADSL supports a variety of transmission speeds, determined by
line quality, wire gauge, and the distance between the customer premises and the
central office.
INDIA- MARKET FORECAST, 2001- 05 |
||||||
DSL Modem |
||||||
2001 | 2002 | 2003 | 2004 | 2005 | CAGR | |
DSL Modem (Units) |
4820 | 6459 | 9329 | 13856 | 22504 | 36% |
% Growth |
34% | 44% | 49% | 62% | ||
DSL Modem (Revenue Rs M) |
54.22 | 61.35 | 74.63 | 99.76 | 130.52 | 19% |
% Growth |
13% | 22% | 34% | 31% | ||
DSLAM |
||||||
2001 | 2002 | 2003 | 2004 | 2005 | CAGR | |
DSLAM (Units) |
184 | 248 | 356 | 529 | 865 | 36% |
% Growth |
35% | 44% | 49% | 64% | ||
DSLAM (Revenue (Rs M) |
133.4 | 161.2 | 195.8 | 259.21 | 371.95 | 23% |
% Growth |
21% | 21% | 32% | 43% | ||
Source: IDC India |
On other hand VDSL, which supports both the symmetric and asymmetric data
transfer, can be used for high bandwidth enterprise applications. In fact, ADSL
itself in combination with the integrated access devices can be an attractive
broadband option for business users.
IP verses ATM Battle
Interestingly enough, the IP verses ATM battle seems to have entered the DSL
arena also. While the world’s largest DSL equipment vendor Alcatel is still
not ready to believe in IP, UTStarcom, the largest vendor of IP DSLAMs, is sure
that it is the future. "We don’t believe in IP DSLAM because quality of
service is not guaranteed on IPDSLAM, as IP networks are still not reliable.
Today, DSLAM means ATM because we would still need ATM from the CPE to the DSLAM,
as IP would not work there," Alcatel’s Bernard Grave says, explaining why
IP isn’t suitable.
Grave, however, doesn’t discount IP completely and Alcatel has a DSLAM that
supports IP. "We have a universal DSLAM (A 7300) that supports all DSL
types be it ADSL or VDSL and every technology from fibre to IP and has built-in
broadband RAS function that can provide remote connectivity from a central DSLAM.
This is because we believe a service provider will have to consider these many
options if it has to offer DSL service," he says.
Comparison of different DSL technologies |
||||
DSL Type |
Symmetric/Asymmetric | Loop Range (in 000 feet) |
Downstream (Mbps) |
Upstream (Mbps) |
IDSL | Symmetric | 18 | 0.128 | 0.128 |
SDSL | Symmetric | 1000% | 154% | 154% |
HDSL (2 pairs) |
Symmetric | 12 | 1.544 | 1.544 |
ADSL (G.lite) |
Asymmetric | 1800% | 150% | 26% |
ADSL | Asymmetric | 12 | 6 | 0.64 |
VDSL | Asymmetric | 3 | 26 | 3 |
Asymmetric | 1 | 52 | 6 | |
Symmetric | 3 | 13 | 13 | |
Symmetric | 100% | 2600% | 2600% |
Ruchir Godura of UTStarcom contends otherwise and believes IP is the future.
"Before IP DSLAMs came in, ADSL was only good for plain Internet access and
not high-speed applications like video streaming and TV over IP," he says.
His company, which leads the market for IP DSLAMs, has been shipping heavy
volumes of its AN- 2000 IP DSLAM platforms (it won a $130 million contract with
Yahoo Japan in 2001). IP- based DSLAMs (as opposed to ATM based DSLAMs) have
gained significant momentum in the DSL market, especially in Asia Pacific. The
spread of networks with Ethernet and IP technologies (that would allow for the
delivery of new IP services such as VPNs, VoIP, and VoD) has been the main
reason for this growth in deployment of IP-based DSLAMs (in place of ATM-based
DSLAMs). "IP/gigabit switching technology is the mechanism being adopted by
the big operators in Japan, Korea, Taiwan and China due to enormous
cost-performance advantages it has over ATM. In Japan, our IP-DSLAM is being
used to deliver 4 Mbps of unconcentrated bandwidth to residential subscribers,
enabling them to receive two channels of broadcast-quality video over their ADSL
lines," Godura claims. UTStarcom recently signed contracts with China
Telecom to deploy approximately 80,000 lines of its IP-based ADSL broadband
solution.
IP or ATM, the fact is that DSL has been one of the bright spots in an
otherwise dull telecom business across the world. And it is very much likely to
remain the broadband technology of choice across the world at least in the
medium term. This is largely because among all broadband technologies DSL is the
one that requires the least amount of investment in existing infrastructure and
can be easily deployed.
What all this could mean to Indian SPs?
Even though no concrete initiative has been taken yet by any service
provider (including the incumbents) in India to offer DSL services to their
fixed line customers, a market surely exists. To begin with, fixed line
operators can look forward to upgrading their dial-up Internet subscribers to
DSL. This would, of course, happen only if subscribers get a DSL service at a
cost that is cheaper than what they pay while using a dial up.
While today there are DSL equipment available that can bring any kind of
voice, data or video service to home or business users, they would not be needed
in India at this point of time. This is because the prime concern of most Indian
users is basic connectivity–something that is cheap, fast and reliable.
Moreover, most Indians are not still ready to pay for content intensive services
like video-on-demand or TV-on-demand. May be, that would be the next step. Also,
from a service provider’s perspective, DSL could emerge as an important
competitive advantage especially for the incumbents as they can provision voice
and video services as well. Also, with competition from the wireless services
(imagine a scenario wherein fixed line subscribers begin shifting to wireless
alternatives like CDMA even for Internet access because that’s better than the
64 kpbs dial up), the fixed line providers would be under more pressure to do
value-adds to the plain telephone connections. Even though fixed operators like
BSNL and MTNL may be inhibited by low and widely dispersed PC penetration to bet
too much on DSL services, they could at least make a beginning. After all, if
broadband cannot be taken to Indian users with a technology like DSL (which
requires far less investment in the existing infrastructure both at the service
provider and user end than any other alternative) then it cannot reach them with
any other alternative either.