Tackling the needs of cellular users inside a building can mean more revenue for operators — whether from enhancing service for a corporate customer or ensuring shoppers stay happy by being able to compare prices. But the technology and planning can make a crucial difference to service. Steve Gold reports.
In-building cellular coverage differs markedly from coverage outside, not least because of the way in which radio signals propagate inside large buildings.
One of the biggest issues facing installers when planning in-building coverage is the type of calls being made — whereas callers on, say, a motorway, will tolerate transient network busy signals on outbound calls, users in, a shopping mall, for example, are far less tolerant.
This is because motorway users tend to be business people who must make their call and, if they travel often enough, they will be aware that calls can — and do — fail for a variety of reasons. They will usually hit the redial button without thinking and the call will succeed, usually on the second or third attempt, even in traditionally high call volume areas such as busy motorway intersections and traffic queues.
In the shopping mall environment, the bulk of outbound cellular calls tend to be from leisure users who are usually driven to call someone on a spur of the moment basis. If the call fails, for whatever reason, there is a strong chance that the call will be aborted in its entirety and the mobile phone user will delay making the call until they are in the vicinity of a landline, or when they are near a payphone.
The only exception to this rule is the growing breed of off-peak contract and pre-pay users that can make calls for free or at very little incremental call costs. There is a strong chance these latter types of callers will use their redial buttons, but the downside here is that the incremental revenue for the network operator is far less than for calls originated during business hours on the motorway.
Corporate systems take off
When digital cellular (GSM) first arrived on the scene a decade ago, major companies used private wires into their choice of cellular network in a bid to cut the costs of outbound calls, as well as the cost of staff calling into the office using their mobile.
Whilst this had the added advantage of allowing staff to have an PBX extension number allocated to their mobile, it still meant that office-bound employees had (at least) two phones on their desks – one fixed, the other mobile.
These days, thanks to the rapid growth in out-of-town business parks and industrial estates – something the cellular carriers could not have envisaged when they designed their networks – coverage in many office buildings is equally as patchy as in shopping malls.
For this reasons, the networks are busy commissioning in-building coverage systems for major companies, allowing staff to reliably make and receive calls whilst in the office, as well as laying the foundations for next-generation PBX services such as transparent number remapping and automatic handoffs between mobiles and fixed lines.
AlanDick already has a frame agreement in place with one of the UK mobile carriers and is currently busy installing between 60 and 70 corporate in-building systems a year.
It’s worth noting that many corporate schemes are the result of operators offering sweeteners to persuade a client to sign up to their network. The sweetener guarantees a higher service level to the client, so requiring the installation of a dedicated in-building system.
Thanks to the falling costs of the technology, the networks have a choice of systems to install in or adjacent to the company site — either a distributed antenna system supporting two to four GSM channels, with each channel supporting up to eight simultaneous voice calls; or a macro base station capable of handling even more channels.
A typical micro base transceiver station (BTS) that feeds the distributed antenna system comes in a case with a footprint slightly larger than a typical VCR and costs around half the cost of a macro base station installation.
Despite the extra costs associated with a macro base station, the units have the advantage of being expandable by simply slotting in an extra card or two. With micro BTS-fed systems, the networks can only daisy-chain complete systems together, so reducing the overall flexibility of such a system.
And with the increasing use of high-speed mobile Internet services by cellular users, macro base stations are increasingly finding favour, since a base station GSM channel supporting eight simultaneous voice calls can often only support two active GPRS calls.
In a busy office environment, it only takes a few GPRS calls to swamp a distributed antenna system. A macro base station, on the other hand, can be configured to cope with such call traffic.
Call blocking examined
According to Colin Garrett, a commercial engineer with AlanDick, the cellular and allied transmission technology specialist, the level of call blocking — defined as the percentage of call requests refused by the network due to lack of resources — is around 2.0% on a general outdoor GSM network.
“Indoors, it’s more likely to be around 0.2%, because, as network engineers, we have to be careful to ensure a higher degree of success for calls inside offices, shopping malls and major public buildings such as railway stations and airports,” he said.
Garrett says there are unique problems facing planners wanting to design and install an in-building network.
“There are a number of issues that need to be addressed. In years gone by, in-building coverage was satisfied by a combination of mini-base stations — known as pico-cells — and leaky feeders,” he explains.
Today, most in-building network infrastructures centre on the use of distributed antenna technology and only rarely are leaky feeders used, despite their relative cost advantages.
The problem with leaky feeders, says Garrett, is that it is very difficult to control where the signal goes. With a distributed antenna system — essentially a series of antennas distributing the same GSM signal at much lower power levels — the propagation of the signal in various directions can be carefully controlled.
Distributed antenna systems also have the advantage that they allow mobiles to be handed off to the outdoor network as soon as the user reaches the exit/entrance of the building in question.
This is a crucial consideration, says Garrett, since it means that the in-building system need only handle call traffic to/from mobiles physically located inside the building itself, rather than also handling traffic from mobiles adjacent to the building.
“It’s important, because this type of traffic can be handled a lot more cost-effectively by the outdoor network, and helps engineers keep the costs of indoor systems as low as possible,” he says.
Indoor signal propagation is not always about cost issues; often it’s about safety, as Garrett explains:
“We had one installation on a couple of adjacent rigs in the Middle East where it was desirable to be able to use a mobile phone inside the various rig buildings, but totally inappropriate to use a handset whilst outside on the rig’s deck,” he says.
“The use of mobile phones on an oil or gas rig is very carefully controlled for safety reasons, since a spark from a mobile phone can have disastrous consequences. There was one very real instance we heard of where a mobile’s clip-on battery triggered a gas explosion on board a rig, so this incident reinforced our existing safety-first approach when designing for the use of mobiles in the presence of volatile gases or liquids,” he adds.
Using a leaky feeder on an oil or gas rig would result in GSM coverage all over the rig. Using a fibre-fed, distributed antenna system, with carefully planned propagation, can limit the signal to within the building itself, whilst providing no coverage outside on the rig’s deck.
Installing distributed antenna systems can work out to be more expensive than using leaky feeder technology, but the potential returns on investment — and the consequent payback times involved — can be quite healthy.
Garrett recalls the time when a significant number of workers were on an oil platform waiting to make calls back home.
“When the in-building network went live, there was a surge of calls made by these workers, with a significant increase in call revenues for the network operator concerned. The net result was the cost of deploying the on-platform cellular coverage was all but paid for before the commissioning engineer had completed his final checks and left the platform,” he says.
Future challenges
“Planning and installing indoor coverage requires a distinctive set of skills for the engineering staff concerned. The job requires considerable knowledge of signal propagation issues on the part of the project manager, whilst installation engineers need to intimately understand structured cabling issues,” says Garrett.
Interestingly, the last few years has seen a subtle change in the types of calls made from large indoor settings such as offices and shopping malls. Increasingly, users are making more and more mobile Internet calls from these environments, mainly thanks to the support for this facility on a growing number of handsets, and a significant reduction in the cost of such calls.
Operators such as H3G (3) and Orange have been offering bundled options for mobile Internet calls to their contract customers, whilst Vodafone has been bundling a few pounds-worth of free data and allied service calls in with its contract calling plans for the last 12 months or so.
This has led to a surge in the volume of data calls – whether GSM data or GPRS – made from indoor mobile phone users in recent times. According to Garrett, this has resulted in GPRS becoming a key issue for in-building coverage, meaning that the volume of network resources available to users in large indoor spaces has had to be significantly increased.
With the expectation of seeing multiple 3G network operators in the UK and other European countries later this year, Garrett predicts that indoor network coverage planning will become even more complex, especially with 3G’s 2.2 gigahertz waveband being sandwiched between the GSM 1800 megahertz frequencies used by the likes of Orange and T-Mobile, and the growing popularity of the 2.4 gigahertz waveband used by public WiFi network operators such as BT OpenZone, Swisscom Eurospot and others.
“And that’s before we even start to get into the issue of private WiFi networks operating on an in-building basis,” he said, adding that the task of a network engineer working on in-building coverage issues is getting more complex as the months roll by.