As UMTS finally takes off across Europe, network planners are exploring how best to meet projected 3G subscriber growth. The technology at the base station RF interface holds many of the answers, writes Joerg Springer
In the latter half of 2004, the third-generation (3G)/universal mobile telecommunications system found its feet in Europe. This technology should provide answers to many of the business challenges facing the region's aging second-generation (2G)/global system for mobile communications networks. While Western Europe's GSM networks still enjoy steady subscriber growth, many are over a decade old, and face serious capacity limitations.
UMTS is founded on wideband code division multiple access (W-CDMA) technology, and promises relief from the current capacity headache. W-CDMA offers a capacity-per-MHz far greater than that of time division multiple access (TDMA)-based GSM, plus reduced OPEX. It also promises more established and sustained average revenue per unit growth -- a powerful driver in an industry climbing its way out of a three-year ARPU slump. The earliest experiences with 2.5G data services suggest that the more advanced 3G will be an important factor in industry growth.
Although the European '3G beast' is now flying, there are unique challenges ahead from a network expansion and RF perspective. Today's 2G-to-3G transition is a different scenario to that of the mid-nineties leap from first generation analogue services to digital.
In just over a decade, Europe's cellular services have matured dramatically, with penetrations at around 85 per cent. The downside of this is that all the prime base station sites are occupied. The environmental requirements regarding site location and visibility have also 'matured' to become some of the world's most demanding.
The transition from TDMA-based GSM to W-CDMA-based UMTS technology also influences network planning. Where TDMA planning strategies are based on minimising co-channel interference by re-using a select number of channels over a group of cells, W-CDMA uses the full frequency band in each cell. Moreover, W-CDMA cells are said to 'breathe' -- the size of the cell varies with the number of callers within the cell, the transferred data rate and so on. The resulting co-channel interference that can occur in the W-CDMA network increases the noise floor, and progressively depletes the capacity of the network. It presents a notoriously tougher network planning challenge when compared with GSM, particularly in addressing the interference resulting from adjacent cells.
Perhaps most challenging of all is subscriber expectation with regards to quality of service (QoS). No longer are subscribers willing to condone dropped calls and fades -- Western Europe has arguably the highest cellular QoS in the world. The new UMTS services have much to live up to.
The majority of Europe's urban cell sites are rooftop-based. Given the tough site acquisition conditions, the easiest 3G roll-out option (and the one largely chosen to-date by Europe's UMTS operators) is co-siting.
The situation on European rooftops today has much in common with a crowded early morning commuter train -- no-one enjoys the congestion, there are established long-term disputes and rivalries between some 'passengers', but on the whole, the system works. To accommodate UMTS spectrum, new antennas are required, so the 'train' needs to be reorganised. The most popular strategy adopted to-date is the multi-band antenna solution. This is manifesting in strong demand across Europe for dual- and tri-band antenna solutions supporting combinations of UMTS 2100 MHz, GSM 900 MHz and GSM 1800 MHz.
A further challenge is co-siting interference. When antennas operating at different frequencies are located in close proximity, there is potential for RF interferences. These are caused by intermodulation products orspurious emissions, which can in turn lead to BTS or Node B blocking. The most extreme cases of these occur when the core band spectral separation is narrow (a pair of UMTS 2100 MHz and GSM 1800 MHz services is a most obvious case), and the antennas are physically close. As a result, UMTS/GSM co-location isn't always straightforward. In some cases, it simply isn't practical, and the new UMTS operator is forced to opt for a site that is nearby, but 'sub-optimal'. The RF challenge is to make the best of such a bad situation, and to optimise the RF footprint to suit the alternative location.
The upshot of this highly constrained site location scenario -- coupled with the exacting requirements of W-CDMA network planning -- is that Europe's 3G operators are demanding higher levels of base station RF precision and flexibility. First and foremost is the issue of antenna performance: new-generation 'precision footprint' antennas feature diminished side and rear lobe radiation levels, improved null fill, and increased front-to-back ratios.
'Flexibility' is being sought on a number of RF technology fronts -- specifically in the control of cell footprint size, shape, direction and power. To compensate for CDMA-style cell breathing and less-than-optimal site locations, variable electrical tilt is de rigueur. Increasingly, this is accompanied by remote tilt control functionality, linked back to the network management centre via industry standard communication protocols.
There is also a demand for tower-mounted amplifiers (TMAs) across the majority of Europe's 3G sites. These provide amplification in the uplink signal from the terminal, which overcomes losses in feeder and co-siting components, decreases the system noise, plus increases the potential cell size. The need again is for flexibility -- a broad choice of amplification levels, dual and multiband configurations, and a wide selection of antenna gains.
In addition, the challenge of W-CDMA adjacent cell interference has created a demand for alternative apertures. Where the 65-degree tri-sector is the norm in 2G/GSM networks, antenna apertures of 90-, 45- and even 33-degrees permit the W-CDMA network planner to 'break the symmetry' of the final cell pattern, and thus minimise cell-to-cell interference.
The not-too-distant future holds even more RF challenges -- significantly, the evolution from a coverage-driven to a capacity-driven strategy. In the very short term, we'll see UMTS operators continue to expand and enhance their 3G coverage across the major city and urban centres. These are the areas that present the greatest revenue earning potential, and are also the most voice capacity-challenged.
Longer term challenge
More challenging, though, is the longer term. Analysts predict a 50-fold increase in Europe's UMTS subscriber levels between end-2004 and end-2009. In essence, this suggests Western Europe will see 3G subscriber levels rise to almost equal those of the region's current GSM count, in a time frame just over half that afforded to the evolution of GSM. This represents extraordinary subscriber growth, and presents unique challenges to 3G network planners and RF technology providers.
To meet the fast-growing capacity requirements as GSM subscribers migrate to UMTS, we can expect to see even more advances in base station RF technology. These will almost certainly be founded on two key elements: advanced 'hybrid' (a mix of active and passive components) antenna solutions, plus greater intelligence and control functionality built into the antennas.
While today's antennas are purely passive, tomorrow's antennas will need to integrate active RF conditioning components, such as low-noise amplifiers, multiplexers and filters, with even greater levels of control. Similarly, by providing more onboard intelligence within the antenna, an increasingly broad range of antenna pattern parameters might be adjusted and controlled. This will provide superior levels of flexibility to the network planner.
Over time, we'll also see a vast improvement in W-CDMA network simulation tools. This should result in more dynamic and intelligent network management strategies, and possibly lead to the realisation of the so-called 'dynamic antenna performance control'. Here, the adjustable parameters of the antenna components -- both active and passive -- could be corrected in response to the simulation tools, perhaps even in a closed-loop real-time configuration. It is these types of super-flexible base station RF solutions that will play a significant role in the establishment of UMTS in the longer term.