Code red

Code redEurope may be lagging the US in the requirement for software radio for personal communications, but its research activities suggest it could have more capable products. Dr Walter Tuttlebee explains
Software radio promises personal communicators that change their function simply by downloading new code. It is a concept which is attracting significant research attention in Europe even though the requirement for software radio is more immediate in the US.
The successful roll-out of GSM digital cellular radio has resulted in users taking for granted the ability to use their handsets for voice and data around Europe and elsewhere in the world. The US, in contrast, is a more recent convert to several competing digital standards. This has led to an obvious requirement for multimode handsets – one that could be met by a software defined radio handset.  
 
Similar opportunities arise for software basestations in the US, supporting AMPS and the newer digital standards. However, with one notable exception, suppliers of such equipment have failed to find markets for their products. Moreover, apart from concerns over intellectual property, third generation mobile systems in the US have had a low profile to date.
The European requirement for multi-mode handsets and basestations is less immediate. The hoped-for demand for dual-mode DECT/GSM products has been slow and this market, when it does emerge, will be small compared to cellular.
However, what software radio does promise Europe is a way of tackling the evolution of basestation infrastructure from GSM to UMTS. For handsets, it offers products capable of handing off to GSM in those areas where UMTS roll-out has yet to reach.
You might think the market differences will result in commercial software radio products emerging in the US first. Yet work in Europe promises to pave the way for more capable, powerful products needed for third generation systems. The key question influencing the commercial evolution of software radio is where can it add value?
European mobile operators are continually seeking ways to differentiate their services to gain market share and retain subscribers. Downloading of applications is one way of achieving this. Limited applications download has been possible using the GSM SIM toolkit. Already a range of added-value applications such as traffic and weather reports, wireless banking and online gaming have been developed.
Smartcard manufacturers are also supplying both GSM SIM cards and bank credit cards. Combining the SIM card with the electronic purse for online purchases is clearly imminent.
Smartcard flexibility will further advance with Java. Manufacturers have developed cellular phones capable of running downloaded Java software, and Cellnet has already conducted one such product trial. Such Java-enabled smartcards offer handsets the potential of enhancing their functionality on a standardised and ongoing basis, allowing network operators to quickly introduce services and applications. Experience gained with these products and applications will provide valuable experience for future software radio concepts involving the download of code that reconfigures the air interface.
What next?
If there is no commercial motivation in Europe for developing GSM software radio products, the same cannot be said for the third generation UMTS standard.
UMTS promises a range of mobile multimedia services – from voice and data to videoconferencing and Internet access. Supporting this range requires considerable flexibility in terms of data rates, bit error correction schemes and the enabling of both circuit and packet switched services.
In turn, the complexity of the emerging UMTS terrestrial radio access (UTRA) standard is much greater than current generation mobile systems. This new market with its demanding requirements is set to give signal processing developments a spur, much the way GSM did a decade ago.
One notable European project to better understand the requirements for software radio for UTRA is FRAMES. This project has been developing air interface technology, including the wideband CDMA and time division/CDMA (TD-CDMA) systems adopted for UTRA by the European telecoms body ETSI.
As part of its development activities, FRAMES has estimated the baseband processing complexity required as between 500 and 2,000Mips.
Another European project, FIRST, has examined many aspects of software reconfigurable air interface implementation. Four demonstration multi-mode testbeds have been built – two configured as mobile stations and two as single-channel base transceiver stations – to demonstrate software radio implementations of DCS1800 and a simplified 1.6 MHz TD-CDMA air interface as a proof of technology feasibility.
Some 230Mips processing power is needed to support eight-slots of this simplified TD-CDMA air interface; additional processing would be needed to implement the emerging ETSI UTRA specifications.
Looking at future RF requirements, work is underway in the PROMURA project to develop a prototype programmable wideband RF sub-system for UMTS. This will accommodate operating frequencies from 500 to 2,500MHz, with 6MHz channel bandwidths. Dr Walter Tuttlebee is business development manager for radio communications at Siemens’ Roke Manor Research, which is active in both UMTS and software radio technologies. This article is based on a longer paper “Software Radio Technology: A European Perspective”, published in the February issue of the IEEE Communications Magazine, Volume 37 No.2.  
 
A key objective of this work is to drive semiconductor technology development, in particular exploring the process impact with respect to BiCMOS and SiGe technology. Optimum digital processing architectures for soft terminals is also a key topic of research. An emerging trend for high complexity DSP systems is a move away from the single powerful processor towards a combination of DSP processor with a reconfigurable hardware processing sub-system, enabling the offloading into hardware of certain processing-intensive operations.
Such issues of hardware reconfigurability for UMTS, for both terrestrial and satellite access, are being looked at as part of the SORT project. This aims to differentiate between critical real-time and common functions to determine the performance/ implementation complexity trade-offs.
Basestations typically use more complex technology than handsets, while at the same time having more relaxed power and size constraints. As such many of the issues addressed above apply equally to basestations. DSP-based solutions which are beyond handset use due to power dissipation constraints may find earlier application in basestation implementations. Indeed, early software radio basestation products developed in the US to address today’s multiple standards use such approaches.
Software radio is still in its infancy, but it is set to have a profound impact on the personal communications industry. It has made its transition from an academic idea to commercial relevance. The advent of UMTS in Europe will serve as a focus for the next transition, from commercial relevance to market-place viability.


Leave a Reply

Your email address will not be published. Required fields are marked *

*