No. 286 (March 2001)

UMTS: Mobile Communications for the Future

This book gives a down-to-earth, relatively non-technical, description of today's most highly-hyped technology: UMTS – Universal Mobile Telecommunication System. UMTS is part of the 3rd generation of communication systems (3G), sometimes called IMT-2000 in ITU circles.

As its name implies, the UMTS system has been designed to become a universal communication system, potentially superseding all other means of communication. The first letter in "UMTS" stands for Universal but could equally mean Ubiquitous or even Unique or Uniform. The uniformity has been achieved at the service level and allows the use of different technologies in different parts of the world, in a flexible manner.

The book was prepared by experts at CSELT (Centro Studi e Laboratori Telecomunicazioni) in Turin, Italy, under the editorship of Flavio Muratore. This version of the book has been translated from Italian but, unfortunately, the English used is sometimes a bit clumsy and the sentences are too long. Also, the reader is liable to lose the thread at times, because too many acronyms appear in the text. On the other hand, the numerous close-ups were found to be very useful, as they give more detailed information on specialized subjects.

The content of the book covers all the major elements of the value chain: service requirements, radio access techniques (in broadcast terms – channel coding and modulation), network access and infrastructure, and terminals. A chapter on the role of satellites in mobile communications is also included, while a section on service-quality testing outlines the test set-up arrangements but does not actually contain any validation results for the UMTS system. The last chapter titled "Research Topics" discusses some interesting topics that require further R&D work, such as space diversity multiple access (SDMA) and "smart" antennas, software transceivers and off-air downloading.

The main technical and operational features of UMTS – which is a versatile but rather complex system – are very well described. It may be useful to briefly highlight them in this review. UMTS uses W-CDMA (Wide-band Code Division Multiple Access) technology. The channel bandwidth is 5 MHz and each operator uses a pair of channels, one for the uplink and one for the downlink (i.e. 2 x 5 MHz). All users served by the same operator share the same frequency channel (this feature is thus the same as in OFDM-based SFN digital terrestrial broadcasting) and are "separated" by assigning a different "code" or sequence to each. Each sequence is multiplied by a factor, so that the resulting "chip rate" is much higher than the information rate (the chip rate is set to 3.84 Mchip/s). The sequences assigned to different users are selected such that the correlation between them is minimized.

Simultaneous users cause mutual interference, so the more users there are means more interference. The system's capacity is limited by the level of mutual interference. Interference levels can be reduced by using power control at the base station, but this technique may be less efficient at high terminal speeds. In addition to its well-known roaming capability (used in GSM), CDMA allows for soft hand-over which is made possible by the microdiversity mechanism. The latter implies that a mobile station, while moving across different cells, can simultaneously be linked to several base stations (nodes B). The book gives an excellent and clear description of the roaming mechanisms.

It is interesting to note that the useful bit-rate of the UMTS system can vary from 144 kbit/s up to 2 Mbit/s, depending on the service requirements and the mobility characteristics. Unfortunately the book does not provide a more detailed insight into the relationship between the bit-rate available and the terminal speed. It would be interesting to consider the BER (bit error ratio) at different C/N (carrier-to-noise ratios) but such information is not included in the book, although it is a fundamental characteristic of any communication system.

A major novelty of UMTS, when compared with GSM, is its capability of carrying not only voice communications but also multimedia services. To this end, ATM Adaptation Layers (AALs), and IP transport protocols to carry data in packet mode, are used. Voice services will still use switched-circuit technology (as in GSM), but multimedia services will use IP packet-switched technology. While UMTS is basically a point-to-point connection system, it is also capable of providing common user data services in a broadcast transmission mode (i.e. a point-to-multipoint connection), either in transparent or unacknowledged mode (i.e. without guaranteeing the packet delivery). It should be pointed out that the broadcast mode may not be commercially viable for the operator, and much too costly for the user. The book does not give a cost analysis for  the UMTS system when operating in the broadcast mode. It would be interesting to compare the cost of UMTS and broadcast services delivering the same television service to the same number of viewers – ranging from several tens to several hundreds of thousands of simultaneous users.

Any new wireless communication system needs spectrum, and so too does UMTS. The book briefly touches on spectrum-related issues, stating that the bandwidth available for the terrestrial component in Europe is 155 MHz (i.e. from 1920 to 1980 MHz for the uplink and 2110 to 2170 MHz for the downlink, as well as from 1900 to 1920 MHz and from 2010 to 2025 MHz for an "unpaired" part of the spectrum). Depending on the commercial success of UMTS services, further spectrum may be required in future. It would not be surprising if any claims for additional spectrum were to target some chunks of spectrum from today's broadcasting bands below 1 GHz, as these are most convenient for mobile reception. The book hints that more spectrum may be required in the future but does not indicate where it would be possible to find it.

The book is very optimistic about the market demand for UMTS services and the likely speed of market penetration in Europe. According to the book, UMTS is planned to roll out in Europe in 2002. In the light of the huge market success of GSM and the current transition towards the GPRS system, many industry experts are questioning whether the year 2002 is still realistic for the launch of UMTS. They are already questioning the tens of billions of dollars spent by mobile phone operators on 3G technology. For example, Vodafone Group PLC paid nearly £6 billion in Britain, and Deutsche Telekom AG paid Euro 8.5 billion in Germany, for 3G licences. These billions were spent merely on acquiring the frequency spectrum that was auctioned off for UMTS by the national governments. This expenditure on the spectrum is judged by many as being so large that the building up of the network infrastructure, the construction of the management control and transaction centres, and the production of attractive services and applications may be severely delayed or even hampered.

There is a fundamental question whether 3G networks are at all viable and even necessary, as the recent advances in data compression software may make 2G networks capable of offering many key services that 3G promises to offer – at a fraction of 3G's cost. Today's 2G systems can handle e-mail and most other applications easily. For example, the wireless Internet service run over 2G by Japan's NTT DoCoMo has been successful at bit-rates as low as 9.6 kbit/s.

In addition, there is growing pessimism, widely echoed in the technical press, about the technical capabilities and the quality of services offered by 3G systems, as well as the high complexity (and cost) of commercial UMTS terminals and administration / control mechanisms which operate the whole system internationally. There is also the question of backwards compatibility of terminals: it is not clear when inexpensive, lightweight, handsets that can run on both 3G and 2G networks could be available. Nokia is saying however that 3G terminals will be available in 2002. It should be pointed out that the main advantage of 3G will be to overcome frequency congestion in zones where GSM is already saturated. There is also a question mark hanging over the coverage of 3G networks. The licences typically require that operators cover "most of the country" within a seven-year time limit. In a larger country, such as Germany, the implementation of that kind of coverage may cost Euro 5 billion.

The UMTS system is being promoted in the European market by the UMTS Forum (http://www.umts-forum.org). The EBU has established a working contact with the Forum and has started to discuss several important matters of common interest – in particular, future UMTS content and applications, interoperability of broadcast and communication networks, spectrum-related issues and some issues concerning the convergence of user terminals. For example, UMTS could be used as the DVB/DAB return channel. On the other hand, DVB/DAB could be used as a point-to-multipoint extension of UMTS to carry multimedia services to the general public. In other words, UMTS and digital terrestrial broadcasting systems may be able to work together (convergence!). This area requires further study – from the technical, commercial and legal/regulatory viewpoints.

In summary, this book gives an excellent introduction to the key technical and operational features of the UMTS system.  However, it omits to give the results of laboratory and field tests that would allow the reader to gauge the extent to which the system will fulfil its performance expectations in real-life environments (for example, indoors and in high-density urban areas). Even so, the book offers a very useful introductory text for EBU engineering and management staff who are keen to learn more about the UMTS system.

UMTS: Mobile Communications for the Future
Flavio Muratore (Ed.)
Hardbound volume of 249 pages
Ref: ISBN 0-471-49829-7. Price £39.95.
John Wiley & Sons, Chichester, UK, 2000.
http://www.wiley.com/

Franc Kozamernik

(back to top)

ADSL, VDSL and Multicarrier Modulation

This book gives a very comprehensive and detailed overview of the different DSL (Digital Subscriber Line) technologies – from the physical layer point of view. DSL has been designed to resolve a "last mile" problem. It will bring multimedia services and the Internet to the consumer at home, using the existing, unshielded, twisted-pair copper wires that are used for the "plain old telephone service" (POTS), while allowing for the existing voice communication to continue. The book covers the theory and practice of the modulation, channel coding, data capacity, duplexing methods, spectral compatibility and system management of DSL. It does however provide a short summary of higher layers of the DSL systems – such as the ATM end-to-end network architecture and a summary of the protocol stacks used for media and internet transmission over telephone lines, but its main purpose is the description of the lowest (physical) layer.

The book is written by an American expert who spent most of his professional life in developing, testing and promoting different communications technologies. This reviewer had the privilege of meeting and talking to Mr Bingham at an ITU-R meeting in Los Angeles in the early nineties. At that time he was working for Amati and developing an OFDM (Orthogonal Frequency Division Multiplex) system for a variant of the IBOC (In-Band On-Channel) digital audio broadcasting approach. In retrospective, however, his efforts have not lead to fruition – as the USA is still stalling, many years later, on the adoption of a viable digital radio system for terrestrial delivery. Nevertheless, Mr Bingham was quite successful in developing a DMT (Discrete MultiTone) system, a sort of a simplified OFDM system which, in the meantime, has been standardized for asymmetric DSL (i.e. DSL) by the ITU in Recommendation G.992: it may soon be adopted also for very-high-speed DSL (i.e. VDSL). He was so successful that Texas Instruments (TI) bought Amati in 1998. Even more, in recognition of his work, there were some rumours that TI would change its own name to California Instruments. In the event, after some argument, it was Amati's name that changed: firstly to AmaTI, then AmaTI and finally just TI.

This book is not the usual type of textbook. It is something between an academic textbook and an engineering handbook, and is aimed primarily at practical, yet specialized, design engineers. It is relatively too detailed for ordinary broadcast engineers who not directly involved in design and research on DSL systems. It is interesting that the author likes to use mathematics very much, with quite frequent "excursions into more exotic realms" (his words). I particularly like his sections entitled "Unfinished Business" in which he openly reveals the unresolved issues or the issues that are still pending. These sections show undoubtedly his high degree of open-mindedness and self-critical spirit. Similarly, the author often introduces "caveats" at the beginning of the section to "warn" the reader that the section lacks a peer-to-peer review and it might be flawed or even technically disputable. In reality, these are brilliant engineering pieces that simply could not be critically reviewed by other EBU experts, because of a lack of time, but they seem to be technically sound in general (as far as this reviewer could judge).

The book gives the fundamentals of multicarrier modulation (MCM), and its simpler variant called DMT, and compares it with single-carrier modulation (SCM) in terms of distortion, spectral efficiency and latency. The consideration of the effects of channel distortion and the side-lobe problem associated with DML makes for very interesting reading. One of the challenges is how to achieve an "optimal" trade-off between different engineering parameters.

At this moment, ADSL technology is gaining in popularity throughout Europe. In most countries, it provides a capacity of up to 512 kbit/s downstream and at least 64 kbit/s upstream. Its capacity critically depends on the distance from the switching nodes. Theoretical values for the data rates are up to 6.8 and 0.64 Mbit/s for downstream and upstream, respectively, within a radius of about 4 km. Within 6km, the bit-rates go down to 1.5 and 0.176 Mbit/s, respectively. ITU G.992.2 defines a simple version (i.e. ADSL lite) with a wider range of data rates and ranges.

The book describes in detail very high-speed DSL (VDSL). This technology will be used primarily in hybrid fibre/copper systems to connect optical network units to customer premises. VDSL ranges vary from 300 m to 2 km and the corresponding aggregate (down plus up) data rates vary from approximately 58 to 4.6 Mbit/s.

It should be pointed out that DSL technology is not the only one to transport two-way high-speed data to and from a customer's premises. Other technologies include coaxial cable, optical fibre and the wireless technologies such as GPRS, UMTS, wireless LAN, Bluetooth and others. There is no general answer to the question of which of these is the best or the most suitable from the economical, technical, financial, social or environmental points of view. The battle between these technologies has already started and it will be interesting for broadcasters to watch which portion of the market share each of these telecom technologies will conquer. Broadcasters are increasingly becoming agnostic about transmission technologies – they are going to use any available technology to bring their content to the general user – provided that it meets their expectations in terms of quality, cost and reliability.

ADSL, VDSL and Multicarrier Modulation
John A.C. Bingham
Hardbound volume of 289 pages
Ref: ISBN 0-471-29099-8. Price £60.95.
John Wiley & Sons, Chichester, UK, 2000.
http://www.wiley.com/

Franc Kozamernik

(back to top)

European Broadcasting Union Case postale 45 Ancienne Route 17A CH-1218 Grand-Saconnex Geneva Switzerland techreview@ebu.ch