Extending the wireless reach within

As the call for greater mobile communications coverage rings out around the world, wireless distributed communication systems are continuing their invisible and inexorable advance.

The ability to access mobile telephony within the world's artificial complexes - be they shopping malls, high-rise buildings, university campuses or underground train stations - is a relatively new and unappreciated development.

Wireless was once a phenomenon largely of the outdoors; the visible signs of its enabling infrastructure were the tower-mounted antennas that rose above the landscape. Today, wireless distributed communication systems (WDCS) bring confined-coverage solutions to places previously unreachable by radio signal. The result is that inside the longest tunnels, largest buildings and deepest mines, clear reception is now commonplace.

The situation is evolving. No longer are WDCS solutions limited to special sites, today they have become pervasive throughout the world. Consumers now expect reliable services wherever they go, but behind the scenes there is much work to be done to make this a reality. A combination of careful planning and innovative RF technology is providing the answers.

Wiles of confined coverage

For wireless technology group Radio Frequency Systems (RFS), the most important facet of each WDCS solution begins at the 'point of interconnection' (POI), where each mobile operator connects its services to the radio infrastructure. From here, a chain of unpowered 'passive' and/or externally-powered 'active' components are linked together, culminating in lines of radiating cable or point-source antennas to provide the crucial WDCS RF interface.

Given the vast range of needs and the daunting variety of WDCS solutions, the technical choices facing operators can be confounding. "There really is no one solution fits all," admits Chris Wilson, WDCS Area Product Manager RFS Asia-Pacific.

To compound the situation, the complexity involved with designing and implementing these systems is growing. David Turkington, General Manager of Comsill, who has provided a variety of in-building distribution systems in Hong Kong, believes that there is a growing pressure on operators.

"The requirements of coverage are getting tougher and tougher. Whereas in the beginning, people were happy with basic coverage in public areas, now they want coverage in the fire exits and even in the toilets with the door shut. With the scope getting greater, the law of diminishing returns is beginning to set in - you have to double the size of the antenna system just to cover a few fire-exits and stairwells," says Turkington.

Co-ordinated combining

Extended coverage-area demands are just the tip of the iceberg. The increase in the number of multiple-operator or 'shared' systems is creating its own issues. The first of these is the so-called 'combining problem', caused when too many new operators try to add their services to an existing site.

Initially the approach to multi-operator environments was to allow each to implement their own RF distribution system. In places like Hong Kong, where there are six major operators, this led to difficulties, as Turkington points out. "While the first or second may be welcome within the building, the later arrivals may not get such an enthusiastic welcome. It means more antennas, more cable space and more hardware in general," he summarises.

Allowing more people to share the existing system is not necessarily an easy fix. "Many of the base-stations combine transmit and receive. If you want to seed a new operator into a shared antenna system, the first thing you have to do is strip the signal into separate transmit and receive ports. So what you have is a system in which signals are combined within the base-station and uncombined outside - not only is this superfluous, it is very inefficient as well," says Turkington. Ultimately, the combining problem constitutes a planning issue for operators. On the design side, careful attention is needed to combine the different cellular networks at the head-end of the antenna system. Significantly, explains Turkington, the filtering and combining needs to be done in such a way as to ensure that there is no mutual interference.
The RFS response to such challenges has been embodied in its 'head-end' signal combiner assembly. Comprising two families of combiner units: the broadband 'BBC series' distributive combiner coupled with band-specific 'MCC series' multi-channel combiners, the assembly allows as many as 16 different operator signals to be cost-effectively driven into a single distribution system.

Quest for coherence

Throughout Asia, confined coverage systems are being upgraded to accommodate third generation (3G) wireless, and control of interference has become a critical design challenge. While coverage may be acceptable, the presence of interference in a system - be it carrier interference in the downlink (manifested as 'intermodulation') or 'noise' on the uplink - will have a limiting effect on data-rate.

"It's not just a matter of the coverage but rather the quality of the coverage," remarks Wilson. "People consider that noise is normally narrowband and should not concern 3G too much - that's not strictly speaking the case, when you consider some of the higher-order intermodulation products."

"For joint-operator systems, noise in the upband is potentially a very serious problem. Given that each operator's performance will be judged on their performance in bandwidth, unchecked interference could play the devil's hand with 3G."

According to Turkington, there is no facility in 3G to isolate adjacent sites, and this has implications for the boundaries between indoor and outdoor wireless systems. "If you have frequency re-use in a 3G system, the overlap may cause 'alarm' interference. For example, your mobiles that are on a high floor in an in-building system could cause a lot of interference to the external system at many sites. Just by raising the noise a little bit, this can reduce the capacity of the external system over quite an area," he says.

Turkington suggests that 3G WDCS solutions will be most effective when supported by an appropriate frequency plan, in which indoor and outdoor systems are assigned different frequencies. Beyond this, it is incumbent on operators to select equipment and system architectures that keep interference to a minimum.

An active debate

At the heart of the WDCS design philosophy is the need to provide adequate frequency isolation, and it is here that the perennial debate between 'active and passive' continues to unfold. For their part, active solutions, many of which are based on the use of remote amplifiers or fibre optic repeater systems, have developed with the aim of meeting larger and more sophisticated in-building needs.

In Korea, the approach has been to embrace active systems, with WDCS fibre-optic solutions used in both outdoor (to supplant base-stations) and indoor settings. Michael Cho, chief executive officer of Cylux Technologies, believes that understanding the limitations of the fibre-optic medium is essential to making these active systems work.

"In one implementation, Korea Telecom Freetel (KTF) installed a subway solution based on digital fibre optic lines and repeaters, whereas SK Telecom specified analog fibre-optic links and repeaters in their section of the subway. But it's all linked together - as long as they are both able to put out the full band of services," says Cho, who stresses that the variation in systems is not transparent to the mobile user.

By contrast, the detractors of the active, and particularly fibre optic, approach claim that the solutions are dogged by a lack of scaleability and flexibility. "Fibre optic solutions are generally okay for a single operator, but if you have multiple operators, you start to suffer from intermodulation interference problems," argues Turkington. "And if you want to add a new frequency band, then you have to put in a new fibre optic system. Whereas with a piece of coaxial cable, you just put in more wire."

RFS' Wilson generally agrees, emphasising that passive systems are more affordable and easier to install and maintain, but he adds a caveat. "The active and passive question is subject to the project you're working with. Our preference, however, will always be for a passive solution, but sometimes the size of a project or the coverage area forces the designer to use an active system approach."

In the past, fibre optic solutions have often suffered from a reduced dynamic range - a problem that has exacerbated intermodulation in the downlink and noise in the uplink. As one element in its high-end WDCS solution set, RFS has invested much effort in providing an 'RF-on-fibre' solution that specifically overcomes this, thanks to high linearity, low-noise transmit and receive technologies. In addition, the active system features a high-gain remote site amplifier, the 'I-RFU', that is purpose-built to support multi-band and multi-carrier systems.

Departure from duplex

Despite some specific active applications, Turkington affirms that the lion's share of WDCS installations in Asia will continue to be passive. For tunnels, metro systems - and increasingly, in-building - applications, the choice is invariably to employ 'leaky coaxial cable' (LCX), also known as radiating cable, as both the transmission line and antenna.

Faced with the need to accommodate an ever widening range of services, without lowering performance, Wilson says that RFS is continuing to extend the limits of its Radiaflex ultra-wideband radiating cable technology. "The challenge that RFS has been addressing over the years is to continually optimise the cable performance to achieve high-performance broadband," he says.

Both Wilson and Turkington agree that one of the next major steps towards combating interference issues will be the move away from a duplex approach to in-building coverage systems. Operators have, in the past, tended to opt for single leaky coaxial cable (1LCX) solutions - in which transmit and receive signals share the same transmission and distribution systems.

The growing problem of passive intermodulation (PIM) in shared systems, however, has inspired many to seek alternatives. "Passive intermodulation is the process where different carriers' signals interact in the connectors and junctions of dissimilar metals, albeit at a very low level," explains Wilson. "For a single-cable solution, there is no isolation between the transmit and the receive band. So, even if the intermodulation is really low, it can still cause significant performance degradation in the uplink path, and the effects of PIM are not able to be filtered out."

Although more expensive, solutions based on two leaky coaxial cables (2LCX) - with separate lines for transmit and receive - now offer clear advantages, including reducing the effects of PIM on the end-to-end system performance.

"1LCX may be suitable in the short term, but in the long term it raises some serious doubts," comments Wilson. "What we see from our experience in Hong Kong, Singapore and Australia is that 1LCX was implemented initially and it worked very well. Later, due to various reasons - be they ageing or the introduction of new services - the performance deteriorates to a level that is no longer acceptable. Inevitably an upgrade to a 2LCX solution occurs."

There are clearly cost-benefit issues at stake. Not every operator can afford immediately to convert to a 2LCX system, and it is not the only way in which gains in performance can be made. Wilson revealed that RFS is continuing to carry out research and development to minimise PIM in its products. Furthermore, the company provides a full spectrum of solutions for WDCS: from the relatively simple and cost-effective 'low-end' products, right through to the 'high-end' solutions, particularly useful for safety- and high-performance critical systems.

"For instance, in the metros, there is vibration and the environment is very acidic and harsh," says Wilson. "Because of the dust metal in the air, components can corrode. With this corrosion, you get the 'rusty-bolts' effect - which is one of the main causes of passive intermodulation. That kind of application will always require the high end."

Impending convergence

Even as operators continue to work to accommodate more operators and services, new WDCS issues continue to emerge. Turkington believes that one area that will present interesting challenges will be the combination of wireless local area networks (WLAN) with 3G.

"There is a potential for intermodulation being caused within the frequency bands utilised by GSM, UMTS and WLAN," he says. "There are no WLAN/UMTS networks yet but, in the future, if people throw up access points independent of the antenna system, they could suffer interference; it will need to be carefully integrated."

Beyond current technological offerings, Cho believes that the ultimate goal is the seamless interconnection between wireless and wireline - a solution known as wireless private branch exchange, or simply 'wireless PBX'.

"If you are only providing mobile wireless WDCS solutions, you are only fulfilling 50% of your customers' telephone requirements," he emphasises. "Wireless PBX means that you can get rid of desktop phones in a business. You have a unit to talk to the PBX and control the wireless side as well - it is the ultimate WDCS solution."

Whatever eventuates, it is clear that wireless systems are set to continue their indoor evolution for some time to come.