Technology for Competition in Broadband Services

Disclaimer

The views expressed here are those of the author and not necessarily those of the ANU, the ACS or other bodies the author is associated with.

Introduction

This is background material prepared for the Inquiry into competition in broadband services, by the Australian Senate². It was prepared at short notice two days before the hearings were to commence and so only covers some issues lightly.

This document concentrates on references: (b) impediments to competition and to the uptake of broadband technology; (c) the implications of communications technology convergence.

What is Broadband?

The Australian Competition and Consumer Commission's 2002 survey³ defines broadband as “... any high speed connection greater than 200 kbits/sec over a mix of media”. As The ACCC points out, this excludes PSTN dial up connections that run at up to 56 kbit/sec and ISDN at 64 or 128 kbit/sec.

A broadband service can be supplied by copper or or optical fibre cable, as an adjunct to a cable TV service, satellite, or ADSL on conventional phone lines.

While the ACCC excludes services under 200 kbps from the definition of broadband, the author uses Transact's service in Canberra at 100 kbps. This provides a more than adequate service for home and micro business use⁴. If properly priced, near-broadband mobile telephone based data services could also provide a good service. For users who have no reliable Internet connection, the priority is a quaility connection, not a very fast one.

Wireless

Several forms of terrestrial wireless technologies have the potential to provide broadband. One group of technologies are the IEEE 802.11 wireless standards (a to g)⁵. These are best known in the popular press for “WiFi” short range wireless devices used in coffee shops. However, there are a range of standards for public use:

• IEEE 802.15 PANs Personal Area Networks (10m range),

• IEEE 802.16 MANs Metropolitan Area Networks (km range),

• IEEE 802.20 High Velocity MANs (for vehicles at up to 250 kph).

A public broadband service might use a combination of the standards, either to form a fully wireless network, or in combination with a wired component. Cable can be used to deliver broadband to within a few hundred metres of homes and then wireless for the last segment to the home. Alternatively wireless can be used for tens of kilometers to the home and then a separate wireless or wired link from the roof mounted antenna into the home. In any case this requires an infrastructure of antennas and transceivers to be installed and is intended for urban areas.

Mobile Phone Based Wireless Data

Another form of wireless broadband is based on upgraded mobile telephone technology. Current Australian mobile carriers supply so-called “2.5G” technology (such as GPRS for GSM mobile telephones). This supplies a service of 53.6 kbps, similar in speed to a dial-up connection. The equivalent CDMA service from Telstra provides up to 144 kbps⁶.

Pricing the Issue for mobile data

In theory wireless mobile telephone based near broadband services are a good way to provide services quickly to a dispersed population. Only a minor upgrade to the mobile telephone network is required (or has already been done) and no cabling to the subscriber;s home is required. However, telcos have priced the data services for mobile telephones as a premium service for executive business people sending short messages. Tesltra's lowest proce for GPRS data is 0.8 cents per kbyte (in their $85 for 10 MB per month plan⁷). This is more than 50 times the rate charged for Telstra's ADSL (wired) BigPond Broadband service (no session fee and $0.159 per Mbyte)⁸.

BushLan

Both the IEEE 802.11 and mobile telephone derived wireless standards are designed for dense urban environments with large numbers of subscribers. They require transmission towers every few kilometers (or in some cases every few hundred metres). Alternative technologies for sparsely populated regional areas are technically feasible, but have not be developed due to the lower likely revenue generated. The Australian National University is developing such a system, called BushLan, which shows promise:

The Research School of Physical Sciences and Engineering (RSPhysSE) and the Faculty of Engineering and Information Science (FEIT) of the Australian National University (ANU) are developing a new wireless technology - the Bush Local Area Network (BushLAN).

Each LAN extends around a regional centre equipped with a base station to the Internet backbone. BushLAN allows a one hop route to connect remote users.

We are aiming for a minimum data rate of 100 kbps (higher data rates are feasible).

From Why BushLAN?, ANU, 2002, URL: http://wwwrsphysse.anu.edu.au/bushlan/background.html

Data rate not the main issue

While the speed of the data service is most often discussed with broadband, it is not the most important issue with providing an Internet type data service to the general public and small business. More important is to have a continuous, reliable, low latency connection which is not connection or time charged. Technologies which provide similar speed to a dial-up connection, but a better quality connection may provide a better public investment than high speed services.

Dial-up Internet services require several seconds for a connection to be made and tie up the telephone line while in use. The user has to pay for a telephone call each time they use the Internet and have to limit their use due to time charges.

The quality of service on a dial-up connection can be unreliable as a different circuit is used for each call. For web browsing the theoretical speed of the connection is less important than the latency of the connection: that is how long between the user clicking on the link and the data starting to arrive in response. Latency is a problem usually on satellite based services, but some mobile phone data services can also be slow.

So called “Broadband” services are not inherently more reliable or faster than dial up services. Last week while in Beijing to assist with the Olympic Games⁹ , the author subscribed to a hotel premium broadband service for $US10 per day. This provided about 132kbps throughput for large files, but with long latency and frequent failures. The result was that this service was less usable for ordinary web access than a typical Australian dial-up service in a capital city.

Conclusion

Speed is not the primary issue with broadband services; cost, availability and quality are. Available wireless mobile phone based technologies can already provide a near-broadband service in urban areas of Australia, but access to this is limited by the high charges applied (more than 50 times that for wired services). Internationally standardised wireless technologies provide the prospect of making broadband more widely available in urban areas. New Australian research bring the prospect of affordable services for regional Australia.

About the Author

Tom Worthington is an independent information technology consultant and author of the book Net Traveller. Tom is one of the architects of the Commonwealth Government's Internet and web strategy. The first Web Master for the Australian Department of Defence. In 1999 he was elected a Fellow of the Australian Computer Society for his contribution to the development of public Internet policy. In 2002 the Australian Computer Society elected him an Honorary Life Member for outstanding services to the ACS over an extended period. Tom lectures on electronic commerce and web technology at the Australian National University and is a Visiting Fellow in the Department of Computer Science. He is a current director and past President of the Australian Computer Society, as well as having been elected a Fellow and Honorary Life Member. Tom is also a member of the Institute of Electrical and Electronics Engineers and the Association for Computing Machinery. Last week Tom was an invited speaker at a symposium in Beijing to plan the official website for the 2008 Olympic Games.

Authorised for publication: This submission was accepted and authorised for publication 12 November 2003 and verbal evidence was given that day. I summarised the evidence as "never mind the bandwidth, feel the quality" and a transcript of it is available at <http://parlinfoweb.aph.gov.au/piweb/view_document.aspx?ID=94375&TABLE=COMMSEN>. At the request of one of the Senators the document "IT Industry Attraction Project: Information Technology Enterprise in the Great Southern Region of Western Australia, an IT Industry Consultancy for the Great Southern Region Development Commission, 8 December 2000, was also included in the evidence: <http://www.tomw.net.au/gsr/>. Tom Worthington, 3 December 2003