Though a late starter, the Indian broadband market is ready to hit the big time. Contributing to the growth impetus are the new technologies entering the Indian market and the fact that the country today utilises just one percent of available bandwidth. Gaurav Patra & Shipra Arora track the growth avenues of this technology in the Indian scenario

Currently in a nascent stage, the broadband infrastructure market in India has just started taking shape. It is still some years before our broadband infrastructure matures to the level prevalent in the US or some countries in the Asia Pacific region. However, broadband is now becoming more and more relevant here because of the extent of infrastructure being laid out in the country, indicating a boom time ahead.

What is broadband?
Technically speaking, it is a type of data transmission in which a single medium can carry several channels simultaneously. Cable TV, for example, uses broadband transmission. In contrast, baseband transmission allows only one signal transmission at a time. Today, most communication between computers, including the vast majority of LANs, uses baseband transmission. An exception to this is B-ISDN Network, which employs broadband transmission. However, in general, the term broadband is used by many for high bandwidth usage. “Broadband is defined as high-speed Internet connections that have download speeds of 128 Kbps or more,” explains Shekhar Avasthy, head of Internet Communications and Convergence Research at IDC India.

Typically, broadband users can be classified into two segments: corporate users and home users. Since multiple machines are required in corporates, they invariably use broadband connectivity. But while Net-through-cable has become a reality in some Indian cities, the number of home users using this facility is still very low.

According to Shekhar Avasthy, telcos have started paying attention to the core layer of their network thus boosting revenues in the optical equipment market

Key drivers
Just like the ISP boom a few years ago, a wave of broadband service providers has now hit the market. The players include existing ISPs who have begun providing broadband services, as well as new players like Spectranet, who deal exclusively in broadband. Apart from these, a number of government entities like GAIL, Power Grid Corporation and Indian Railways are also in the race. With so many players in the market, is there enough demand to support all of them?

Before we discuss the issue further, a point to be noted is that the broadband market is basically of two types, terrestrial- and satellite-based broadband infrastructure/services. As far as terrestrial broadband is concerned, there is currently a huge glut of bandwidth in the country. According to industry experts, hardly 1 percent of the total bandwidth available is being utilised. In fact there is excess of bandwidth even in mature markets like the US, where only about 3 percent of the total capacity is being utilised. “This is happening because the service/infrastructure providers are going on adding capacities while there are not many applications to use that much. What needs to be seen is whether there are enough applications to use the bandwidth being added,” says Ashish Chowdhary, vice-president of enterprise sales at Hughes Escorts Communications. Also, the applications in the Indian context are still very preliminary. Take the case of video conferencing, which has not yet matured, resulting in users sticking to 36 Kbps applications. “There is one set of people who don’t feel the need for huge bandwidth, and another set who want more bandwidth but are not willing to pay for it. So cost also becomes one of the hindering factors in the Indian market,” comments Chowdhary.

As far as broadband on a fibre optic backbone is concerned, telephony is the major application. Apart from this there are few other applications which can run on a fibre optic backbone. Indian corporates have already started looking at value-added applications like Internet connectivity, Web conferencing, video conferencing, multicasting, cable TV, etc. “Currently, most of the companies (service providers) are riding on telephony. However, they can also provide all other services very cost effectively,” informs Chetan Choudhari, president of Fibre To The Home (FTTP), Aksh Broadband.

Experts say that basic telephony is one area which has tremendous potential to drive the broadband market. With so many players in the race, the cost of telephony will go down further, which will boost the broadband market. This is because most of the basic service providers are going in for optic fibre. However, it is the Internet and video which will remain the main drivers as these services will fetch higher RoI for the broadband service providers. In a competitive scenario like this, value addition is going to be the buzzword.

Apart from corporates, it is the home and SME segments that are going to drive the broadband market in the metros or A-class cities. In the home segment the drive is going to be through entertainment and other value-added services.

The rural segment is emerging as another crucial market for the Indian broadband sector. However, what is significant and good news for the broadband service providers is that the demand in this market is not for high-end value-added applications as in the case of urban areas. The major applications that are expected to drive the rural market are telephony, Internet, video and various e-governance applications. According to Choudhari, the growth in the rural segment will be even faster since many villages do not have an existing copper network and so not much effort will be required in putting the fibre optic infrastructure in place. Understanding this opportunity, Aksh Fiber Optics has already started laying optic fibre in 20 villages with another 100 to be covered in the next few months.

As far as the infrastructure part of broadband is concerned, the market is already abuzz with a lot of activity. Take the case of investment in broadband infrastructure. If one goes by recent press reports, something like Rs 30,000 crore is being invested in laying optic fibre cable within the country, with the largest chunk of it allocated to the creation of an intercity backbone.

Meanwhile, certain steps need to be taken to boost the demand for broadband services. Of these, the access network needs immediate attention, otherwise this investment in the backbone network shall remain unutilised or poorly utilised. “The lack of return on this capital investment shall be a repeat of what happened in the US. Indeed, the demand for broadband is tremendous, but you can’t see it in real numbers for reasons of non-availability of proper infrastructure,” comments Naseem Ahmad, CEO, Broadband & CATV Networks, HFCL.

Out of sheer compulsion, a few years back Indian Railways introduced computerised reservation. A similar compulsion will force the government to adopt e-governance, tele-education and tele-medicine, and business to take up e-commerce. All this will need broadband, multimedia delivery pipes to homes, offices and wherever one goes. This in turn will create opportunities for broadband players.

SDH/DWDM market
India has primarily been a switching country as far as telecom infrastructure is concerned. However, with the fast spread of a telecom network and the introduction of data communication services in the country, telcos had no option but to upgrade their network to support ever-expanding voice and data services. “Telcos have started paying attention to the core layer of their network. This gave impetus to the optical equipment market, which took off from there,” opines Avasthy. “Today, the optical equipment market in India is at a stage where there is potential of exponential proportions both in terms of unit shipments and value,” he adds. With the entry of multiple private operators in the Indian telecom market, the need for high-speed voice and data infrastructure has really fuelled demand for SDH (Synchronous Digital Hier-archy) and Wavelength Division Multiplexing (DWDM). IDC estimates that the SDH market will grow at a CAGR of 66.5 percent for 2001-2006. On account of a higher adoption rate and lower base, IDC also predicts that the DWDM market will grow at a CAGR of 82 percent for the same period.

Ashish Chowdhary says that under-utilisation of bandwidth is happening because service/infrastructure providers are adding capacities while there are not many applications to use the bandwidth

Optical Ethernet market
Optical Ethernet enables WAN communication services using the Ethernet packet format and Ethernet equipment. The technology runs over all kinds of optical circuits—unmodulated, Dense DWDM and Synchronous Optical Network (SONET). Optical Ethernet allows such standard Ethernet speeds as 10 Mbps or 100 Mbps. However, so far, the basic backbone access speed is 1 Gbps, over which users can buy bandwidth in units of 1 Mbps or 10 Mbps. An Optical Ethernet service can be cost-effective compared with other broadband access technologies like T3/E3, but so far it is limited to areas in which fibre is already installed.

The strategic value of this new way of using Ethernet goes far beyond cheap access. It can be used both for last mile access and in local backbones within service provider networks. This technology can deliver a range of services like high-bandwidth Internet access, frame relay and leased line service, transparent LAN service, and a packet-based wide area data service designed to interconnect separated LANs at a data rate sufficient to carry the full bandwidth of the LAN connected. “Transparent LAN services within MANs are seen as one of the strong early markets for optical Ethernet equipment,” feels Avasthy.

With high-speed access as the starting point, service providers can expand their offerings to MANs, which are usually variable bit-rate services. However, because optical Ethernet is so closely related to Ethernet LAN, there is particular stress on providing full-bandwidth connections among dispersed LANs, known as transparent LAN services. Although there is no reason why optical Ethernet-based networks cannot be national in scope, the bulk of connections are likely to fall within metropolitan areas, where companies tend to have many offices. That is where optical Ethernet services can be offered as MANs—where fibre will be installed both as metropolitan rings and to individual buildings. According to industry experts, optical Ethernet is likely to have a vast price advantage over ATM and other technologies in conveying IP traffic. As per IDC, this market is likely to achieve a CAGR of 43 percent by 2006.

Market trends
It is important to understand that the prime driver for broadband, apart from faster downloads, will be overall increase in productivity and the overall value proposition of being connected. The end user is now slightly more mature, in the sense that he is already exposed to using the basic non-bandwidth intensive applications like chat and e-mail. Corporates are giving much more Internet-related facilities to users, like providing everyone with an independent mail ID and a machine that has Internet access. “Applications like video-on-demand which are bandwidth intensive are still unheard of in the market. Till applications like these become available, broadband will not pick up in the home segment,” Avasthy says.

Ashok Juneja says the trend will move in the direction of a small number of users using more broadband capacity, while a large number of people will be happy with normal capacity

Although the Indian market is currently in a very nascent stage, industry gurus expect that it will grow significantly in future. This will be largely driven by the corporate segment. Other key market segments for broadband are retail and finance. The financial segment is considered a key growth driver because of growth in the ATM segment. “There is going to be large relevance of retail networks. Keeping this in mind HECL is about to roll out a retail network of the size of 3,000 to 5,000,” informs Ashish Chowdhary.

Since the consumer market is not yet mature, the customers who will demand broadband in India will be large companies. It will take at least two to three years for the consumer market to take up broadband significantly. From the overall perspective though, the market is growing. More and more conventional mediums at conventional speeds are coming up, and they will mature and lead the way to broadband. A price fall will also make broadband more affordable to users. Fibre and satellite will co-exist and form the dominant technologies. “The trend will also move in the direction of a small number of users using more broadband capacity, while a large number of people will be happy with normal capacity,” opines Ashok Juneja, CEO of Bharti Broadband Networks.

Present situation
According to a study by IDC, ‘Broadband Equipment Market Forecast and Analysis,’ the Indian broadband equipment market is expected to touch Rs 66.52 crore (excluding satellites) in 2005. IDC also believes that the Indian market for broadband access is one of the most dynamic in the telecom space.

The broadband demand scenario can be looked at from two perspectives—last mile and infrastructure. “If one looks at the first one, the market has not changed much over the last few years, growth has not been very significant, and even the numbers are not impressive,” says Avasthy.

The cable modem and DSL (the two leading broadband technologies) subscriber base would have grown by only around 15-20 percent over the last year. The number of players also remains the same. One or two players (Dishnet being the primary player) are offering DSL, so not much action has been witnessed in this space.

But if we look at the second space—the infrastructure backbone—there has been a lot of activity during the past one year. IP1 licencees like Reliance, Bharti, BSES and Tata Power have invested huge amounts to lay optic fibre cables across the length and breadth of the country. Public sectors giants like GAIL, Power Grid and Railtel are also investing heavily in laying optic fibre cables and connecting various cities. Most of these projects (or at least their first phases) will be completed by the last quarter of 2002. With such huge investments, the scenario will change considerably by the end of 2003.

However, one of the important challenges before the Indian market is the availability of last mile connectivity. Currently, companies engaged in providing last mile connectivity (through DSL, cable) are focusing on the ‘Golden Square Mile’ i.e. areas which have clusters of high revenue clients, especially corporates; the service is not available in most parts of other cities. For example, Spectranet in Delhi has been focusing only on the corporate segment.

Availability of last mile is what is going to finally decide the future of the broadband market. Other issues like affordability for end-users, availability of killer applications (which are bandwidth intensive but of such great interest to users that they opt for it despite high costs) and demand for these applications become important only after the availability of the service is there. “For any revenue model, the application/service needs to have a minimum critical mass, which is possible only if the service is easily available. Also, the price at which an application can be offered is directly dependent on the critical mass,” explains Avasthy.

The reason for the slow take-off of cable despite much higher penetration is that the investment required at the cable operator end to upgrade infrastructure is very high. Besides, the cable operator market in India is highly fragmented with many small players; a huge investment is just not feasible for them. Also, the experience of users who have taken broadband connections from cable operators has not been very good—there have been problems with bandwidth and slow connections. As a result, the market for cable-through Internet has just not picked up.

Naseem Ahmad says that though the demand for broadband is tremendous, one can't see it in real numbers for reasons of non-availability of proper infrastructure

Technologies to look at
To find out whether or not a particular technology suits Indian conditions, one must keep in mind that India is a land of vast geographical area with highly varying climates. Thus, to offer connectivity to remote areas, geographically tough terrain like mountains and deserts, scarcely populated areas like remote villages, and of course the more developed urban areas, different technology modes are offered in the market by different vendors. Some of the more popular technology options include jelly filled cables, integrated service digital networks, optic fibre cables, and very small aperture terminals.

• Jelly Filled Cables (JFCs): With a bit-rate of about 28.8 Kbps, JFCs have limited bandwidth capabilities, which make them unsuitable for networks meant for high-speed transfer of multimedia data. Even otherwise, JFCs are susceptible to noise interference by electrical circuits and also to inducing their own signals into other circuits, leading to what is known as cross-talk. Moreover, some areas, particularly rural, have reported frequent thefts of JFCs.
• Integrated Services Digital Network (ISDN): ISDN, digital public telephone network, is very suitable for moving data. Routed networks are possible using dial-on-demand between routers in different offices. ISDN finds multiple uses like voice, data and basic level video conferencing (up to 15 frames/sec). However, ISDN in different countries use different standards, requiring a translation-processing step during transmission. It is a good technology for small file sizes and networks that are not too extensive geographically, but can be an expensive proposition for a national backbone network.
• Optic Fibre Cables (OFCs): The advantages of fibre optic as a communication medium is manifold. OFCs do not emit radiation, so they are ideally suited for computer applications and environments where high voltages and electrical substations are present. Due to the electrical isolation, the problem of cross-talk is also not present with OFCs. Their immunity to outside electrical interference is also very important as it ensures that network synergies with power grids can be exploited. OFCs have low loss, high bandwidth properties, and can be used over greater distances than copper cables. The high bandwidth available in OFCs allows the transfer of multimedia data. Because of low attenuation rates (in data networks repeaters are required only after every 2 km or so), fibre is ideally suited for broadcast and telecom use for longer distances. Being lightweight and small in size, they are ideal for applications where running copper cables would be impractical. Using multiplexers, one fibre could replace hundreds of copper cables. OFCs also offer higher security of data transfer/communication as compared to co-axial cables; an optical fibre has to be broken to tap the signal—leading to loss of signal—alerting the user immediately.
• Very Small Aperture Terminals (VSATs): The advantages of VSATs are many. First, they offer easy access to areas which are difficult to access or where landlines are expensive or not available or not of good quality.

Availability is another key advantage of VSATs. A point that is often misunderstood is that VSAT circuits are more reliable than terrestrial circuits. Leased lines offer at best 99.5 percent availability (almost two days downtime per year). VSAT circuits can offer 99.9 percent to 99.95 percent, depending on the requirements (99.9 percent = less than 9 hours downtime per year). A third important point which goes in favour of VSATs is price. Today, the cost of a VSAT network is roughly half to one-third the price of an equivalent frame relay network. Transmission costs are not distance dependent, and are also not a function of time on line or volume of data transmitted, unlike terrestrial networks. Flexibility and scalability too are not issues. A VSAT network can be scaled up easily according to requirements. A new site can be added to the network by installing a remote VSAT unit (a VSAT has only one piece of equipment at the far end). Mean time between failure (MTBF) for some of the equipment used is measured in years, so it is very reliable. Satellite networks offer excellent security against unauthorised access. All transmissions in VSAT systems are scrambled in digital format. Gaining access to a VSAT system is virtually impossible without authorisation. Every remote VSAT earth station is controlled and monitored. Due to these reasons, banks and financial institutions predominantly use VSAT systems to carry critical and sensitive financial transactions.

How do they compare?
Of the four technologies, ISDNs and JFCs are clearly not suited for the backbone, given the nature and scale of anticipated data transfer needs and the limitations of the technologies in terms of their ability to carry high-speed multimedia traffic. The choice is therefore between an OFC network and a VSAT network. “OFCs thus form a long-term solution given the huge scale of Internet traffic expected in the future...they will be able to exploit synergies with the railway networks and power grids,” says Avasthy. “Also, interconnection between the backbone and ISPs providing the last mile link will be easier. VSAT networks are comparatively more expensive than FOC networks,” he adds.

Agrees Chetan Choudhari, “On satellite you cannot have so much of bandwidth for video. In VSAT, bandwidth is a limitation. The equipment cost goes high as you grow the bandwidth and it is a recurring expense whenever you download. Also, in case of bad climatic conditions like rain, thunder and dust storms, the connectivity gets bad.” He further explains that VSAT is good when it is used for remote area connectivity and there is no other connectivity available. OFCs, on the other hand, are good as a long-term non-recurring expense. “The cost of fibre has come down by almost five-to-seven times in the last eight-to-nine months. Hence, the initial investment costs of laying down OFC have come down drastically. Fibre has therefore turned out to be a cheaper solution than copper for last mile,” adds Choudhari. These factors have made OFC a popular choice for Internet backbones in most countries, except when geography makes a terrestrial network impractical, as in the case of Indonesia with its thousands of islands.

But Ashish Chowdhary feels otherwise. “From the service providers’ point of view, VSAT has a bit of an edge over fibre in terms of initial investment. There is a huge upfront investment required in setting up terrestrial fibre optic infrastructure. However, in case of a satellite, there is a benefit for service providers as the investments are not upfront.” He adds, “Capacity can be leased as demand grows. From the users’ perspective also they can easily add and scale up capacity. VSAT allows you to start out low and scale up as requirements grow. Satellite also has the advantage of reaching the last mile.”

Whatever the pros and cons may be, ultimately both satellite and terrestrial will co-exist and complement each other, rather than compete. There are certain areas where the two might overlap, but by and large they address two different markets. While the terrestrial medium targets core applications (ERP, connectivity for main offices, payroll and HR applications), satellite targets broadband for mid-tier and end-tier retail applications (retail point kind of applications).

Government’s role
After opening up this sector, the government has started taking other positive steps. For example, it has allowed Ku-band to come in. (Ku-band is much easier and cheaper to install.) The introduction of revenue sharing has also been an encouraging step. This has given a big boost to the overall broadband market. “We will continue to work with DoT to lower the size of the antenna from the current 1.2 metres to 0.75 metres, which is the American standard. This is better because the antenna is easier and cheaper to install,” informs Chowdhary.

Chetan Choudhari says that the cost of fibre has come down by almost five to seven times in the last eight to nine months, hence fibre has turned out to be a cheaper solution than copper for last mile

“One of the most important factors which is considered to be a hindrance by players is the import duty on equipment. The government should do something about it,” says Juneja. This is also a matter of concern for satellite broadband players. “The duties on equipment in the satellite industry are relatively high...companies have to pay 35-40 percent. This is presently a barrier to the entry of technology,” complains Ashish Chowdhary, who however concedes that the barrier will eventually go away because of WTO. Industry experts also feel that issues relating to the ICE Bill should be sorted out as soon as possible. “I believe that when the Cable TV Regulation Bill is passed, that will open up some more issues. The government should take immediate initiatives to sort these out,” says Ahmed. Predicts Deepak Malik, EVP for corporate services at Data Access, “International Internet access costs will tumble if the government breaks the monopoly of access to undersea fibre bandwidth capacity.”

But Chetan Choudhari sympathises with the government. “The government played an important role when they opened up the whole scenario. However, we want the government to be even more open in the rural sector. This sector is very important as almost 70 percent of India’s population lives in rural areas.” The government can indeed play an important role in encouraging companies to provide broadband infrastructure for the rural areas. Also, bureaucratic procedures, approvals, etc. need to be taken care of to speeden the process as technology changes very fast.

What’s in store?
On the terrestrial broadband network front, with more service providers coming in, there is going to be a very tough competitive scenario in the Indian market. Even in the US, many companies have not been able to survive the competition and have been forced to close shop. So what is the logic behind the increasing number of service providers setting up infrastructure? Quite simply, expectation of demand.

In this scenario, the break-even for Indian service providers will take much longer than planned. This will also lead to price cutting in the market, so only big players will be able to survive. Moreover, raw bandwidth will not be enough; players will have to provide a lot of value addition. In the case of satellite broadband, the value addition can be end-to-end communication services to customers, video broadcasting and multicasting, ASP offerings, a bunch of applications which customers can use (CRM, messaging), data centres and managed network services. In terrestrial networks, the value addition can be in terms of providing more capacity for doing video conferencing, bandwidth on demand, etc. These can become the key service differentiators for service providers.

The future of the broadband market in India is still hazy. Infrastructure in general and access network infrastructure in particular is the need of the hour, and the moment these start building up, some experts say the market will explode. However, others feel the glut in capacity may never be matched by demand. The lobby that says broadband usage will explode says content creation and conversion in local languages of different regions of the country is an industry which is all set to boom in the coming years. Application development like distance learning, video on demand, education on demand, home shopping and interactive gaming will all become big businesses and create a multiplier effect on the size of the total market.

The lack of proper infrastructure has so far forced the broadband segment to experience slow growth. However, once things start rolling the subscribers will start coming in too. The extent of this flow—a trickle or a flood—will determine the future of the broadband players. Prices will start falling in due course of time, and that will boost demand even more. Predicts Ahmed, “Since the sky is the limit for applications, I do not see any saturation in the broadband market for at least the next 25 years.”