Rediff Logo Infotech Find/Feedback/Site Index
June 16, 1999


Search Rediff

Internet Unplugged: No cords, no modems, no dial-up, only high-speed WWW from IIT, Madras. Professor Ashok Jhunjhunwala of Indian Institute of Technology, Madras, has developed a wireless system to access the Internet.

The system is cheaper and faster and bypasses the regular telephone networks. Networks, that will become exasperatingly clogged with the increase in the number of Internet connections.

Email this story to a friend.

More importantly, Jhunjhunwala's system does not require a modem for accessing the Internet! Besides, the surfer gets a telephone free to talk on while being logged on.

What's the catch?

Same as the one that came with his earlier invention: An indigenous 'wireless in local loop' telephone system. That too was significantly cheaper than any options available across the world.

The catch then was bandwidth allocation for the system and poor faith in Indian technology. Bandwidth might not be an issue this time round, but lack of faith certainly is.

Jhunjhunwala spoke with Shobha Warrier in Madras about the new technology, its context to the Indian economy and the impossibility of competing with multinational companies when it comes to funding.

He warns expensive technology from the West has only limited application in India and relevant indigenous methods are the only way out.

And this is no fancy talk, Jhunjhunwala has figures, economic and technological, to back his claims:

What is the relevance of wireless Internet access systems in India? Did you begin working on the technology only because it is cheaper?

Yes. In India, we have only 15 million telephones for its 1,000 million people. Even in the urban areas, it costs more than Rs 30,000 per telephone line. If you take 15 per cent as financial cost and another 15 per cent for operation, maintenance and obsolescence costs, a telephone connection requires 30 per cent of the Rs 30,000, or Rs 9,000 per year for the telecom operator to break even, which means Rs 800 per month.

Now Internet sits on top of this!

Unless one can spend Rs 800-900 a month, one really cannot afford a telephone or an Internet connection. This is not affordable to more than 2-3 per cent of the population.

Therefore, we are stuck with 15 million telephones in India and ultimately a few million Internet connections too.

We will never be able to get to 100-200 million connections and make an impact in the country unless we are able to do something about this.

One way of doing this is to bring down the cost to Rs 10,000 per line. Then you need only Rs 3,000 a month to break even and suddenly it is affordable to 150-200 million people.

Can technology help in this? That is what we at IIT, Madras, are trying to answer for the last few years. We are trying to find a total access solution. Here one of the key things to the solution is the wireless local loop. And an equally important role is played by fibre networks for direct Internet access systems, which we have recently developed.

This is said to be a low-cost wireless Internet access system. How did you make it less expensive? Is it because there is no copper in it?

Yes. The total cost will be approximately about Rs 7,000-8,000 as opposed to Rs 25,000-30,000. There is no copper in it and that is one of the main reason for its affordability.

Then we have used the standard CorDECT system. We have succeeded in concentrating the Internet traffic at the street corner remote terminal. These are the other two important reasons why we are able to significantly reduce costs.

It is reported that it is not necessary to have a modem in the new Internet access system that you have developed and subscribers can directly connect their PCs to the Internet. How's this possible?

It is possible because of the CorDECT wireless local loop system and the 'remote access switch' that were developed at IIT, Madras.

The new wall set that we have developed has an RS232 port (the standard serial port interface for a PC) to which a computer can be directly connected.

Because you are directly connected to the serial port you don't require a modem. Using either a 28.8 KBPS asynchronous connection, or a 64 KBPS synchronous connection, a link can be established between the subscriber and the Internet service provider.

We have a remote access switch, an RAS, at the DIU or the 'DECT interface unit' which will concentrate the traffic, that is, the data from different subscribers.

At the DIU exchange, we also separate the data so that the voice data goes to the PSTN directly and the Internet data goes to RAS. And from the RAS, the Internet data gets connected to the ISP.

Could you elaborate on CorDECT wireless local loop?

CorDECT wireless local loop system was developed on an international standard called DECT by the TeNeT group at IIT, Madras, in collaboration with Midas Communications Technologies, Madras, and Analog Devices, USA.

The system comprises a DECT interface unit, which acts as a switch or a 'remote switching unit' connected to the telecom network on E1 lines using either R2-MF or V5.2 protocols.

Twenty base stations are connected to the DIU, either using three twisted pair copper wires per base station, or using an E1 radio or fibre link from the DIU to a base station distributor.

The subscriber is served using a wall set, which is connected, to a base station using digital wireless links as per the DECT standard.

And each isolated base station can serve about 50 subscribers at 0.1 Erlang per subscriber providing a grade of service of 1 per cent.

The range supported is 150-400 meters when the base station is mounted at a few meters height in urban streets and 10 kms in the line of sight situations when base stations are mounted on rooftops or towers.

A 'relay base station' would enable service to subscribers located as far as 25 kms from the exchange. Both the BS and RBS use highly directional high-gain antennas.

As both are mounted on towers, line of sight is also available. Thus, even with the low-transmission power used in DECT, a 25-km link can be established.

The RBS picks up signals from a base station as far as 25 kms away and rebroadcasts it to wall sets within a 10 km range.

The wall set is the subscriber end-equipment. It has a RJ-11 interface to connect to a telephone or fax or modem. The CorDECT wireless in local loop system can provide high quality voice telephony using 32 KBPS digital data transmission to the subscriber.

The wall set is locally powered using 230V but has a built in battery to provide 24-hour standby time when used for voice communication.

How is it possible that the telephone line remains free even when the user is surfing the Internet?

How do you connect to the Internet? There is a PSTN (public switched telephone network) connection and a telephone in your home.

You connect a modem and a computer to the telephone.

There is also a ISP which will take multiple telephone lines, will take modem banks, will take a router and connect the router to the router of other ISPs of other cities through PSTN network or lease line.

All that a subscriber does is dial up through the modem to the router to the ISP to get connected to the Internet.

Here you are using the wired line. The PSTN in India has been designed to serve a peak hour traffic of 0.1 Erlang per subscriber, assuming that a telephone is used at an average 10 per cent of time during the busy part of the day.

While a voice call lasts mostly for a few minutes, an Internet call usually lasts much longer. So, an Internet user offers a load of as much as 0.3 Erlang during peak hours.

As the ratio of the Internet users to the total users grows, PSTN will just not be able to handle the load.

The network will get congested and fail to complete a large number of calls. Then, the quality of the dial-up also varies.

Although it provides 28.8 KBPS occasionally, most of the time, it provides only 9.6 KBPS or 4.8 KBPS.

If the ISP has 100 telephone lines with 100 modems, the 101st connection to ISP is not possible. So, the investment increases rapidly and in a linear manner with the number of customers, an ISP serves.

At times, we sit for hours on the Internet and what does it mean? You are occupying or blocking the line. You are paying not only the Internet charges but telephone charges too, which is Rs 28 an hour. If you are farther away, you may be paying long-distance charges too.

As we are setting up a RAS at the DIU, we are not allowing the Internet traffic to enter the PSTN. As several lines get concentrated at the RAS, you need only one line to go to the ISP. Another advantage we can say is, even though we are giving 35 KBPS now, it can be upgraded to 70 KBPS.


From 35 KBPS to 70 KBPS calls for a software upgrade only. You download new software. That's all.

How is it cost-effective?

I told you about the calls getting terminated at the RAS. Of course, our wireless local loop is very cost-effective. Today we are providing simultaneously, voice and Internet access with no additional cost. That is, at the same cost, I am able to provide Internet access and voice communication.

Will it be cost-effective in the low teledensity rural areas?

Yes, in very sparse rural areas too, it will be cost-effective. To cover the whole area, you need only Rs 17,000 per line. And this Rs 17,000 includes both voice and Internet too at 35 KBPS, upgradable to 70 KBPS. We are covering the whole of Thanjavur district at the cost of Rs 15,000 per line only.

It is reported that your wireless in local loop system was cold-shouldered by the government and you couldn't install the system in India even though you have exported the technology to many developing countries?

It is not true. Who told you that? We have already started installing in several places including Madras, Hyderabad, Lucknow, Bhopal and Rajkot. We have got orders from Delhi and Bombay too.

Is it not true that multinationals are getting preference over your technology?

It is true that they are also installing the technology, but at 2-3 times our cost. Yes, there is a little bit of lack of faith in Indian technology, which is natural. I don't blame anyone. I hope that the attitude will change gradually.

Is lack of faith in Indian technology the only reason that they get precedence over you? Or are there other reasons too?

There are other reasons too. Financing is one major reason. When Motorola comes here with their product, it comes with Exim Bank financing of the USA, with a very good financing term of 6 per cent interest rate.

Second, they have their brand name. There is always a belief that if you give order to Motorola, things will be all right. Then, of course, there is a lack of belief in Indian technology. They believe that multinationals can provide better service.

Then, the policies. It took us sometime to get clearance on frequency. You know about that. There were all kinds of other obstacles too. We have overcome most of the obstacles.

Do you find it difficult to convince people in getting finance?

It is not a question of convincing people. Indian banks give finance at 18 per cent whereas the Exim Bank of the USA funds Motorola. The Exim Bank is ready to give loans at a much lower rate.

Multinationals always have this advantage. We have to overcome all that.

How do you plan to tackle these problems?

We have superior technology and we offer it at a much lower cost. Ultimately, what matters to any financier is lower cost and high returns. We have already got a lot of orders.

Would you be able to produce en masse if you get a lot of orders?

We don't produce it. We are not manufacturers; we are technology developers. We have licensed the technology to Shyam Telecom, HFCL, Crompton Greeves and ECIL and they produce.

So, in India we have four companies producing. In China, there is one company. In Brazil, there are a couple of companies. We have a licensee in France. So, that is no problem at all.

You say that your technology is on par with what the multinationals have brought here...

Not on par. Superior. We provide both high quality voice and Internet service at 35 KBPS at the installed cost of Rs 13,000. Nobody else could provide these facilities at this cost. No one will come close to us.

Do you feel frustrated when multinationals are given preference over your superior technology?

It is a part of life. We know that all these obstacles are part of India trying to stand up. I have always emphasised that a high-cost technology is not viable in India. It is viable to a limited extent in very large conglomerates and it becomes less and less viable as we go to smaller towns and villages.

They (the multinationals) are definitely not interested in these areas except in dumping certain systems on DoT. I don't think they will finance in a large way to install it in rural areas. They will not finance it.

Do you develop technology with India's rural background in mind?

Yes. Multinationals like Alcatel and Siemens do technology development for developed countries. Their focus is the needs of the developed countries like the US, and Japan.

They develop what they need. Their need is not to reduce cost. For example, an installed cost of $1,000 a line is well affordable there. $30 or $40 a month is not a problem there. So, their focus is not reducing the cost but adding more features.

But we are concentrating on India and other developing countries. Our needs are different. We need low cost, affordable solution. We use the same latest developments in the technology that they are using. Except that we look at these technologies in an innovative manner because we need an innovative system essentially to reduce cost and provide good service.

You may ask, are you giving poor service? No. We are giving 35 KBPS Internet access, which can be expandable to 70 KBPS and a telephone at Rs 13,000.

This is not a poor service! It is a terrific service! We are trying to reach every rural area because our focus is different.

If for example, Siemens or Ericsson focuses on our country, they will beat us, as they are more capable and better organised. We will not be able to do anything as good as them. But their focus is somewhere else and our focus is here, in India.

So, do you see your technology plying a major role in India in the next century?

I see it playing a major role not only in India but also in all developing countries.


The Negroponte Switch
Dr Jhunjhunwala explains his breakthrough 'wireless in local loop' and what it means for a poor nation.

The Lone Ranger
Professor Jhunjhunwala is fighting the state which has rejected his invention for quick and cheap phones and opted for a US technology which is more expensive and dramatically less efficient.

Tell us what you think