Home > Research and Market Data > Technology research > Overview > Tomorrow's Wireless World

Converged Communications in Tomorrow's World

Foreword

Welcome to this, Ofcoms fourth report on our Technology Research Programme.

This year we report on a diverse range of projects. The management of spectrum naturally remains a key theme within our research programme. However, the work detailed in this report illustrates how our research reflects the breadth of our regulatory duties. For example, we report on our studies into the communications infrastructure required to support novel entertainment services over the coming 20 years. We also present the findings on a study to better understand the environmental impact of communications systems.

As in previous years, we welcome comments on the research reported, and on the scope and direction of our future plans for further research next year. Your feedback can be directed by email to gary.clemo@ofcom.org.uk^(-1-)

Ofcom does not conduct technology research in-house; rather, we engage the services of a wide range of consultancies, public bodies and academic institutions to conduct research on our behalf. The quality of research continues to be high and I welcome the level of stakeholder engagement, through workshops and interviews, that has guided many of this years projects. We gratefully acknowledge the work of the consortia on these projects and the help received from many of the consortia members in compiling this report.

Peter Ingram
Chief Technology Officer

Executive Summary

Introduction

This is Ofcoms fourth Technology Research Report, providing an overview of the technology research programme at Ofcom during 2008/09. This report presents the findings of the consortia of industry, consultants and academic institutions that have undertaken the technical work on Ofcoms behalf. The views expressed are those of the consultants and are not necessarily endorsed by Ofcom. It is worth emphasising that all of the activities described here are research projects; by their very nature they are speculative and relate to visions of the future telecommunications landscape, rather than current regulatory policy.

Ofcom publishes an overview of technology research on an annual basis to inform stakeholders of findings, to stimulate debate and to gather feedback on both the results and the future direction of the programme. While publication of this report does not constitute a formal consultation, we welcome any observations stakeholders care to make.

Context

Ofcom is the regulator for the UK communications industries with responsibilities across television, radio, telecommunications and wireless communications services. By understanding potential future technology developments, Ofcom can determine how technologies and services might develop and shape regulatory policy accordingly.

Key findings

The research during 2008-09 has concentrated more on policy and our regulatory duties than in previous years, where it looked at emerging technologies. Consequently, it has led to a range of results that have important implications across a wide range of policies, especially in the area of spectrum.

The work on congestion and occupancy has yielded some interesting results about licence-exempt policy, particularly at 2.4GHz, as well as opening up new ways of looking at spectrum data through a UK-wide occupancy database. The environmental work raises some important questions as to the extent to which we should consider environmental issues as part of our existing duties.

This year we have grouped the research projects into two categories, summarised in the following sections.

Communications in society: Entertainment and the environment

There are a number of ways to approach technology research within the communications and media domains. One approach might be to focus on a particular technology, to understand how it might provide a new service or use spectrum more efficiently. An alternative approach might be to focus on the needs of a sector of society or industry and then work towards understanding the technologies that may be required to meet those needs. Both approaches are equally valuable and enable an organisation like Ofcom to understand both the details of technology evolution and gain an appreciation of how new and emerging technologies might benefit the citizen-consumer.

In this category we have grouped together three projects on the basis that we are interested in how communications technology could be used to fulfil a role in society, rather than on specific technologies themselves. The entertainment sector is a major user of communications networks and services, covering the production, distribution and consumption of content. We commissioned two entertainment studies, covering short-medium and long term views.

The first project considered whether there were any barriers, either commercial or technical, in delivering services such as video-on-demand across the open Internet in the short-medium term. The project concluded that core networks could be upgraded to provide sufficient capacity as needed. In the case of extreme traffic demands, the costs to do this might be in the region of 1 3 per household per month on top of existing Internet Service Provider (ISP) subscriptions; however, given the increase in value of the data consumed in the home this increase in price might be acceptable.

Looking to the longer term, we commissioned a second study to assess the entertainment services that consumers are likely to want in the next 10-20 years. We can then consider whether sufficient networks, technologies and spectrum exist to enable these services to be provided. The project suggested that there are two key scenarios which might unfold. In the first, users download most content to a home server and then distribute it around the home. Content is side-loaded onto mobile devices via wireless in-home connectivity. Most video consumption is in the home on large screens. In the second scenario the mobile device becomes the entertainment server. Content is loaded directly onto the mobile via a range of networks and video consumption takes place when mobile or in a range of locations.

In terms of communications resources, we noted that the home scenario can be readily accommodated with existing or planned core, access and wireless networks. However, we observed that the mobile scenario could lead to increased spectrum pressure in around 10 years time as demand builds. This demand might be met, for example, with released military spectrum in the 4GHz band.

The study concluded that either of the two scenarios could unfold with equal probability. It may be possible to predict which scenario is transpiring by monitoring a number of metrics such as the take-up of mobile data. We could then take proactive steps to adjust our spectrum policies as a result, encouraging more rapid release of spectrum across key bands, if needed. This is an important result that enables us to better understand whether demand for entertainment services will overwhelm networks in coming years.

The third project in this category seeks to understand the impact communications networks and devices can have on the environment. Our principal duty is the promotion of citizen and consumer interests, which arguably includes such environmental considerations. In the future, we may be expected to consider environmental impacts explicitly, so we commissioned a study to understand whether this would materially change the decisions we make, particularly relating to radio spectrum.

The project noted that assessing the absolute environmental impact of any network was extremely difficult given the many components, such as base stations, handsets, computer servers, civil works, power requirements, eventual decommissioning costs and so on. Many of these would have a large international component for example, handsets are mostly made outside of the UK and might also be recycled or disposed of internationally.

However, it is much simpler to assess the relative environmental impact of different networks since this substantially reduces the scope of those areas that need be considered. The project included a number of case studies, which showed that for consumer services the environmental impact is typically dominated by the end-user devices and not the network. So, for example, although a satellite broadcast network uses less power than a terrestrial broadcast network, this is only a tiny fraction of the power consumed in the home and hence generally not a key reason to prefer one system over another. Another interesting case study showed that, from an environmental viewpoint, a network that made extensive use of femtocells was preferable to one that achieved the same coverage via additional macro and microcells.

The study shows that tensions might arise between existing policy approaches and greater weight being placed on environmental impacts. For example, using spectrum more efficiently might enable additional networks to be deployed, leading to valuable new services but may increase environmental impacts. Promoting competition through infrastructure deployment may also increase environmental impacts. However, it is not uncommon for such tensions to arise and solutions are typically found by weighting the various approaches appropriately.

Spectrum: Modelling, measurement and future usage

One of Ofcoms major responsibilities is the management of the radio spectrum. Better spectrum management leads to more efficient utilisation and an increase in value for all stakeholders. The consumer, increasingly dependent on wireless and mobile communications, directly benefits from our work in this area. By improving our knowledge of how various parts of the spectrum are actually used we can take steps to ensure that the services to which they subscribe meet the standards of quality they expect and have paid for. The projects described in this section include both practical modelling and measurement of spectrum usage, and an investigation into ways in which spectrum might be used in the future.

We have long been interested in the degree of congestion in licence-exempt (LE) spectrum, particularly the band at 2.4GHz. Measurements from previous years research projects suggested only around 20% occupancy but we were concerned that occupancy might not be a good measure of congestion, particularly for WiFi systems. There is also much anecdotal information that WiFi systems are highly congested in cities and failing to work in certain locations. We therefore commissioned the design of a novel monitoring system based on smartphone devices with global positioning system (GPS) and WiFi capabilities. These were then carried around city centres, including into a number of buildings such as railway stations, where they monitored key beacon signals transmitted from WiFi networks and measured performance parameters. A test network was also built in the laboratory to try to replicate the measurements made.

The measurements showed that WiFi was relatively uncongested, even in those environments where there were many WiFi networks. The protocols within WiFi appeared to generally work well in sharing the available resources. Many anecdotes, such as the congregation of WiFi devices onto default channels, were shown to be false. However, the work showed that where WiFi performance was poor, for example in some London railway stations, this was predominantly due to interference from non-WiFi devices. The worst interference appears to come from video senders devices that send TV signals throughout the house, allowing, for example, all TVs to view the signal from a set-top box. This is a new insight, not noted before, and would explain many of the anecdotal observations. It is also somewhat worrying as video senders appear to be proliferating.

The second project in this category investigated a novel way of measuring spectrum utilisation over a much wider area. Our consultants approach involved deploying measurement equipment on the roofs of the vehicles used by a UK-wide sales force to capture information on frequency bands from 10MHz to 6GHz. The project led to the first tool able to show occupancy across the UK in all of our key bands working in real-time from a data set of over 200GB of measurements. Initial analysis of the data suggests it is possible to extract accurate usage information across a wider range of locations than previously possible.

We are currently half way through a two year project which is working towards unifying the many existing propagation algorithms into a smaller set, a so-called wide-range propagation model. Work in the first year has concentrated on identifying the constituent sub-models that, in the second year, will be combined into the unified model. This year has seen the development of a number of sub-models, including an approach to capture the sporadic interference caused by ionospheric and meteorological phenomena.

The next project in this category investigated an approach to determine whether wind farms cause radio interference to wireless fixed link systems. We commissioned a series of field trials to measure the effects of wind farms on fixed link and scanning telemetry systems. The study found that the most satisfactory method for predicting the impact of wind turbines involves characterising turbines in terms of their radar cross section. The limited trials undertaken by our consultants have established a useful method for performing this characterisation.

In addition to practical spectrum research, we also undertake more abstract and theoretical studies. These typically involve predicting ways in which spectrum might be used in the future. Ofcom has already set in place a number of initiatives to make more efficient use of the spectrum. Such major changes, however, can take years to complete and, in the intervening period, there may be a possibility of increased pressure on available spectrum.

We, therefore, commissioned a complex modelling project, which examines possible demand for spectrum from different services including cellular, short range, broadcasting and fixed wireless access. It uses a wide range of scenarios, some of which are deliberately extreme, in order to understand whether there is sufficient supply to meet these demands across a number of different frequency bands and services. Our consultants modelling factors in known and predicted spectrum release, improvements in technology and detailed models of different types of networks.

The study suggests that there will likely continue to be a need to maintain our market-based approach to spectrum management, in order to ensure that the finite supply of spectrum in the most sought-after frequency bands is channelled to the users and uses that generate the greatest benefits for society. Pressure on frequencies below 15GHz may be mitigated by spectrum awards and on the release of spectrum by public sector users. The latter is subject to decisions by government departments, especially the Ministry of Defence^(-2-). The conclusions of the study are broadly consistent with work carried out for the Independent Audit of Spectrum Holdings on spectrum demand for non-government services 2005-2025 ^(-3-).

Demand for spectrum depends on a wide range of factors, including price, whether set by the market or administered incentive pricing (AIP), the availability of spectrum on the secondary market and release of additional spectrum. For example, if licence fees were reduced, demand would rise and shortages would become more likely. A similar result would ensue if spectrum releases were curtailed. The study suggested that for almost all services and scenarios, the effective supply of spectrum will potentially grow faster than underlying demand, with implications for appropriate policies to be followed on spectrum pricing, trading and release. The study concluded that pressure on available supply was most likely to be experienced in two particular cases. First, under a number of cellular scenarios, spectrum demand will increase more quickly than supply in the next few years as wireless data grows. Secondly, under the most aggressive demand growth, the study predicted that there will be increased pressure on spectrum for short range wireless services in around 10 years. However, as mentioned above, these findings significantly depend on the amounts of spectrum public sector bodies decide to release to the market within the study period.

In addition to these spectrum-related projects, we have also undertaken a project that considers the issues around determining the commercial value of spectrum to a potential buyer or seller. The modelling phase of this project is currently drawing to a close. We expect to publish this research report via our website and will summarise the project approach and findings in next years annual report.

The coming years research

Over the past six months, we have consulted widely, both within Ofcom and externally, on suitable topics for the 2009/10 technology research programme. At the time of writing, ten projects are under consideration:

Communications technologies for assisted living.

In this project we seek to understand which services and underlying technologies will be required by elderly and disabled people in order to enable them to lead a fully inclusive life. We can then predict future requirements on communications networks and act, if necessary, to ensure that there are no technical or regulatory barriers to the provision of these services;

The future of advertising.

Here we propose to undertake a study of the technologies that are likely to be used as advertisers turn to increasingly innovative ways to generate revenue. Technologies for behavioural and personalised advertising are of particular interest, along with the implications for the individuals privacy;

Technology horizon scanning

More generally, we need to keep abreast of emerging, and entirely new, technologies. This project will involve seeking expert judgement on the likely impact of emerging technologies across Ofcoms industry sectors;

Assessing network quality of experience

The most common measure of network performance is speed or data rate. However, there are many other ways to categorise and measure performance and many factors that affect network performance. In the future, a more varied selection of metrics will be required to categorise network performance, especially important given the increasing amount of multimedia traffic. We are, therefore, proposing to investigate how the users view of quality of experience relates to underlying, network-level performance metrics. We then propose to investigate technologies that could be deployed to realise quality of experience levels in a network. Such technologies could be used to make better use of the network link through innovative traffic management and may enable users to get better broadband performance without the need to upgrade to a faster line;

The capacity limits of fibre networks

Fibre optic networks offer the promise of data rates to the home in excess of 50Mbit/s and substantially more in future years as the technologies evolve. It is often suggested, anecdotally, that the capacity of fibre networks is so great, it can be considered unlimited. In reality, capacity will depend on a number of factors, such as network design, the technologies used in network equipment and the quality of the fibre optic itself. We are proposing a study to inform us on these limitations;

Understanding the deployment efficiency of microphones in Ultra High Frequency (UHF) spectrum.

One of the key uses of spectrum in UHF Bands IV and V (470 862MHz) is wireless microphones for programme making and special events (PMSE). This equipment is still primarily analogue, although digital technology is available on the market and increasingly being used in professional settings, such as West End theatres. Digital microphones are becoming available, which are more spectrally efficient, but may suffer from latency between input and output. We are proposing a study to develop a better understanding of the spectral efficiency of analogue and digital wireless microphones;

Study of current and future receiver performance

The cost of TV receivers influences their performance. While they may provide a good quality of reception currently, there is a concern that after the switch from analogue to digital TV is completed, cheaper receivers may be susceptible to interference from new services operating in the UHF band. The proposed study will investigate the cost impact of improving receiver performance to ensure that new services can co-exist with digital TV with no degradation in quality;

Locating devices in areas where GPS is unavailable

Ofcom has previously suggested that cognitive radio devices may be allowed to operate in the UHF bands once the switch from analogue to digital TV is complete. Cognitive radios are able to opportunistically use frequencies which may ordinarily be in use by others. One condition of their use is that steps are taken to minimise the risk of them interfering with existing spectrum users. One approach is to maintain a database of observed radio activity at various locations; a cognitive radio would check the database prior to communicating to be sure that its target frequency is not being used. This approach requires cognitive radios to know their location. GPS is a popular positioning service, but may not always be available at all locations. We are therefore proposing a study to investigate other methods of deriving location information;

Understanding likely interference effects of future power line communications equipment.

There are a number of wired and wireless technologies available for home networking. One approach that appears to be gaining popularity is networking through the electrical mains wiring within the house, with the current generation of products offering data rates of up to 200Mbit/s. These products may interfere with existing spectrum users, such as short wave radio, maritime and aeronautical transmissions. The purpose of the proposed study is to investigate how the technologies used in these products are likely to evolve in the coming 5 to 10 years and to understand the likely interference effects on existing radio systems; and

Improving our propagation models.

Precipitation has a surprisingly profound effect on radio waves, especially those at the higher frequencies that are used for fixed wireless links. If our tools do not accurately model the effects of sleet and snow there is a possibility that we will be overly cautious in our approach to spectrum management, leading to less efficient use of the spectrum. This study will investigate and propose ways in which to improve the accuracy of our modelling tools.