The Impact of Microcell Technology on Mobile Communications
Robin Potter. VP, Consultancy. Mobile Systems International.
msi-world.com
At the Venice Conference that introduced GSM to the world in 1987 Prof. Ray Steele and myself presented a paper on microcellular engineering. To my knowledge this was the first attempt in the world to identify the opportunities that would be created by standing traditional cellular engineering on its head and providing street level 'underlays' of cells with antenna below roof level and coverage of only a few hundred metres. The key potential that we saw was for almost unlimited capacity to be delivered through the very high density frequency reuse that could be created and in many respects was an application of the work that was underway at the time in developing the new digital cordless telephone standards (DECT and PHS ) and built on the work at Bell Laboratories on what ultimately became PACS.
What was essentially new in our paper was the proposal that the microcellular underlay would form an integral part of the wide area cellular coverage environment with seamless handover between the two environments. The mobile phone would operate at near cordless power levels within the microcell environment but would be able to instantly revert to its macrocell, higher power operation when handed over into the wide area environment.
The organisers of the conference were clearly not sufficiently impressed with the practicality of our ideas at the time so our paper was relegated to the poster sessions which were generally the domain of the more academic papers and thus attended only by those dedicated enough to want to discuss these more complex papers with the authors rather than relax in the full conference sessions.
Well things have moved on a little since those early days………..
Second generation digital cellular networks have now achieved widespread deployment around the world with GSM technology forming the principal technology base although in is anticipated that more extensive deployment of IS95 derived CDMA will also be taking place over the next few years. Already in the more mature GSM networks some early deployment of microcells is taking place and much attention is now focussed on this area of cellular network design and implementation.
Two key drivers exist for microcell deployment:
Pressure on the availability/reuse of radio spectrum - this is generally a problem for the GSM Operators, particularly in the more competitive markets where less spectrum may be available to an individual Operator.
The need for ' special coverage ' solutions - particularly, initially, to address the coverage requirements for public gathering areas such as shopping malls, railway/subway systems and sports arenas where coverage is not readily provided by the wide area cellular infrastructure.
The Microcell as a Capacity Enhancer
The application as a capacity enhancer alone generally requires the microcell to provide street level coverage over a relatively short range of typically a few hundred metres and, most importantly, for the coverage and hence interference potential of the cell to be significantly constrained much beyond its normal coverage range. Generally this is effected by mounting the cell antenna below roof height - illuminating the cars and the people rather than blasting away at the horizon as is traditionally the situation - and using relatively low radiated power (typically 1 watt or less at the base station).
Since the coverage area of such a microcell largely follows the outline of the streets around its location then a much more irregular coverage exists than provided by an overlaying macrocell. Essentially the microcell is putting capacity into just the area required for high density provision - the area being typically less than 1/10th that of even a high density urban macrocell. With 1 or 2 radio carriers deployed then the capacity, in terms of Erlang per square km that can be delivered by the microcell can also be significantly higher.
Frequency reuse is also claimed to be much better for microcell deployment however this needs some care since all is not what it may seem…. Certainly where a small number of isolated microcells are deployed with no possibility of coverage overlap, then the frequency group assigned to a microcell may be reused by all other microcells in the area. However this is unlikely to be a practical situation for long and as more microcells are deployed, and particularly once overlapping coverage becomes necessary then more normal rules or frequency reuse have to be applied.
Performance optimisation and a good understanding of the practical issues generated by using microcells are also an important elements of microcellular engineering. A vehicle travelling at 20km/hour in a city can transit a microcell at a junction in less than 10 -20 seconds and although this is unusual in our more crowded cities at peak times, it is clearly important that such 'fast' moving traffic is maintained at the macrocell level rather than handed off into the microcell. This is most readily achieved by parameter setting that only permits calls to be originated in the microcell. This arrangement, however does not necessarily capture all the slow moving traffic for the microcell that is desired and more sophisticated algorithms for managing the relationship between the microcell and its neighbours are generally necessary. The important factor here is the application of algorithms that steer traffic from slow moving handsets onto the microcell to maximise the utilisation and in this respect the use of the cell reselection capabilities of GSM are highly effective.
In order to plan effectively the deployment of a microcell layer within a city and to manage the subsequent performance optimisation it is essential that effective tools are available to the engineer to visualise the location options and thus coverage areas of the microcells. In addition the composite coverage of the macro/microcell layers needs to be available to the engineer together with interference prediction and analysis capability in order to plan and optimise performance. It is particularly important to an Operator that initial microcell deployment into their network is done against a longer term plan/scenario - if this is not the case then it is likely that some microcells will require relocation in the longer term with consequent cost and disruption impact on the network.
Not only does MSI's PLANET provide all of the above features, its sophisticated mapping capabilities can enable the Marketing and Engineering functions within an Operator to work together to plan the microcell deployment and to review plan scenarios and options within the office.
An important element for Operators in planning for microcell deployment will be the use of high definition street and 2 or even 3 dimensional databases of buildings of at least those areas of the cities where microcell deployment is contemplated. Such databases can now be readily generated from satellite data although for the (typically 1 metre ) accuracy required for building outline data then higher resolution data as provided by aircraft photography is still required.
Microcells for Indoor Coverage
The indoor coverage application is in some respects more challenging than the outdoor microcell since each building generally requires its own tailored solution to provision of coverage. One useful element of indoor coverage is that, in general, the fabric of the building will tend to contain the radio signal simplifying issues of interference and frequency planning with the wide area network.
The present market for microcell indoor deployment covers 3 main types of environment, namely :
Large area public gathering places such as shopping malls, airport terminals and railway stations. Where these are not readily served by the outdoor infrastructure then the coverage can generally be delivered using 1 or 2 cell sites each with their own local antenna or, if necessary a limited number of antenna with the RF being delivered by coax or now, more increasingly, using fibre with RF/fibre conversion equipment at each end.
Medium/large business premises where the cell site(s) provide the campus coverage, frequently for a corporate package delivered by the cellular/PCS company. If the building is a multi storey block then coverage would generally be provided with antenna or possibly radiating cable (leaky feeder ) systems on each floor with again the RF being delivered by coax or fibre systems to the floor level radiators.
Subway systems where extensive coverage of concourses, platforms and tunnels requires a multi-faceted approach whereby each environment is subjected to its own special needs. In particular the tunnels generally require radiating cable systems with repeater amplifiers at regular intervals and in call handover becomes a requirement because practical capacity /coverage dictate the use of a number of base stations to provide service across the complete infrastructure.
Technology and System Developments
The early deployment of microcells for capacity/coverage enhancement used conventional macrocell equipment. Typically this would be a single cabinet of large refrigerator dimensions but with lightweight feeder cables and low profile antenna (frequently panel antenna can be used) since RF performance is not generally a limiting factor. Now however the key GSM equipment vendors have microcell equipment available supporting 1 or 2 carrier operation at relatively low power within a housing of typically small suitcase dimension with a weight that allows wall mounting to be used. This equipment opens up a new dimension of flexibility for locating the microcell equipment and also minimises site costs and problems of locating equipment in locations secure from vandalism. Most vendors also now have equipment which can operate in tropical environments without any special cooling or environmental requirements and this is an important consideration for flexible deployment.
The next development now underway would appear to be the first realisations of a true 'picocellular' technology base station. During the last year the European Telecommunications Standards Institution (ETSI) has produced the specifications for an exciting new product which is being called the GSM Home Base Station. This will be a compact (ie laptop size) base station which, while supporting a (slightly modified ) GSM (900/1800) radio air interface, is designed to connect to a telephone line and thus enable the GSM handset to operate at home as a cordless telephone connected to the public fixed telephone network. The first home base stations are expected to appear later in 1998 and further, PBX connected picocell products are anticipated to follow on. At first sight the utility provided by having cordless phone functionality in a common handset might be questionable, however once a home base station has the signalling functionality to communicate the location status of a customer to central network servers then this opens up the prospect for Operators to deliver a powerful range of fixed/mobile integrated services including automated call redirection and integration of messaging services. The additional opportunity for the mobile operator to use the home base station /mobile functionality to route long distance and international traffic (regulation permitting) via their mobile network, using the fixed network simply as an access mechanism has also been identified.
The Future is Microcellular
It is tempting to get carried away by the potential that micro/picocellular technology offers to the industry. Clearly the potential is massive as I have tried to illustrate in this article and it is possible to imagine a mobile future where all buildings have their own micro/picocellular environments engineered in a manner that, to the customer, appears integrated with the outdoor world.Large numbers of outdoor microcells would then provide almost unlimited capacity in other traffic hot spots.
The problem with such an idealised scenario is that many technical and economic issues are raised and it may take some time before we have solutions to these.
An outdoor microcell typically supports 2 carriers compared with the 12 to 15 that would be typically found on a macrocell base station at full capacity (assumes the Operator has 2 times 10MHz of radio spectrum). At today's equipment costs the capital cost per transceiver deployed is typically 50% higher for the microcell equipment from vendors at the present time - in future this gap is expected to narrow but this will take some time. In addition the site rental and leased line costs for the microcell are higher on a per transceiver deployed basis - typically 100% higher is not unusual when all factors are taken into account. Thus it can be seen that widespread deployment of outdoor microcells can raise network costs, both capital and operating and at the present time, care needs to be taken that the cost/benefit balance is properly understood.
The indoor microcell has a similar problem but in some markets this is being compounded by the position taken by site owners of public buildings who expect substantial rental returns and in some cases a proportion of call revenue derived from the site. This may not be unrealistic since it is a commercial world within which we live but it is unfortunate that costs may be artificially inflated and development inhibited through these demands.
The situation will be different for the private micro/picocell environments where it is generally in the interest of the site owner to have the equipment and indeed the owner may be expected to fund or at least contribute to the cost of the equipment. The cost of equipment may be factored into the overall call /monthly charge package for the owner. One interesting example is where a site such as an airport may provide both public and private service within its coverage area.
The microcellular environment requires planning tools that can support the high definition coverage predictions and manage the parameters necessary to optimise the balance of traffic and performance between the microcell and macrocell layers. Within the city areas where microcells are planned for deployment then high resolution building datases will be needed - typically to an accuracy of approaching 1 metre accuracy. Such an accuracy cannot be achieved today from satellite imagary however aircraft photography provides a solution.
While the future is indeed microcellular, like all technology it is important to understand where it can be most effectively used and that it is not seen as a universal solution to the challenges of coverage, capacity and performance optimisation that are facing Operators today. With such an understanding, and the support of adequate tools and databases for network planning then microcells offer an effective means of delivering capacity and coverage to deal with crtitical areas of an Operators infrastructure. |
