Digital Advanced Rural Testbed (DART)

Digital Advanced Rural Testbed (DART) is a project initiated by the Technology Strategy Board (TSB) in the UK. Avanti will lead and coordinate the project and provide a senior project manager, experienced in trials and demonstration projects. A project management plan will be developed at the start of the project to provide detailed planning for the project and define project processes. Industry will lead the activity to build the DART testbed. Academic support will be provided by the School of Engineering University of Aberdeen (UoA) and University of Lancaster (UoL).

The DART project will provide a testbed that offers an advanced infrastructure representative of the access technologies that will play a significant future role in serving rural communities using a mix of satellite, terrestrial wireless and community fibre schemes. The testbed provides a unique, opportunity for businesses to experiment (in a safe and collaborative environment) with rural broadband, and its associated efficiencies (e.g. Multicast cost-savings) and unsurpassed coverage opportunities (e.g. Greater potential market for new services). The main goals of the DART Project are to make this advanced infrastructure available, engage content and applications providers to experiment with its innovative features, support those experiments and encourage cross fertilisation of ideas between those experimenting with the services.

Overview

High speed service delivery to unreached communities is recognised as a serious technical and financial challenge and one faced by countries worldwide. Delivery of Internet services to the whole of the UK population (and in particular the "final third") is a key objective for the UK government. This project focusses on piloting video-centric applications to exploit Next Generation Access (NGA) services. Recent consultations on NGA provided by organisations such as the Broadband Stakeholders Group (BSG) and trade association Intellect, have affirmed that a mix of fibre, satellite wireless, and terrestrial wireless technologies will be required to deliver NGA at an affordable cost.

The Digital Advanced Rural Testbed (DART) will provide an advanced infrastructure representative of the access technologies that will play a significant future role in serving rural communities using a mix of satellite, terrestrial wireless and community fibre schemes. The advanced testbed infrastructure will enable experimenters and content providers to evaluate new advanced network features. DART will be constructed using a combination of satellite, terrestrial wireless and community fibre schemes, all of which are expected to play a significant future role in serving rural communities. The testbed will include approximately 200 end-user locations. These will be divided between two sites:

• Site A will be located in Scotland, supported by enhanced direct-to-home or shared-access two-way satellite broadband services.
• Site B will be in and around Wray, Lancashire, with users connected via hybrid satellite and cable or wireless mesh.

The testbed provides a unique opportunity that will allow experiments to be performed with real end-users and advanced network applications and services to explore business cases for commercial service deployment. Experimenters can take advantage of a set of advanced network features expected in next generation broadband networks, especially:

• Multicasting
• Dynamic variation of the Quality of Service (QoS) and bandwidth
• Advanced network caching techniques.

DART also enables exploration of key enhanced service mechanisms, including: metered billing (by time of day, type of data, volume of data), and charging for a premium for uplift of QoS, or connection speed.

Example digital applications that can utilise the features of this testbed include:

• Combining multicast and caching, using satellite multicast to support cost-effective user demand for high bandwidth services and increasing consumption of video (following the iPlayer model).
• Utilising adaptive quality of service to deliver increased capacity on-demand for high quality telemedicine video conference, or to allow payment of a premium for high-speed upload of business-critical files. This includes the ability to charge a premium for uplift of QoS, or connection speed for e.g. telemedicine or video delivery.
• Business models based on micro-payments, allowing content providers to pay for improved delivery of their own content. Support will be provided for metered billing (by time of day, type of data, volume of data). This could include using a multicast Push Video on Demand service and charging content providers per view.
• A flexible service model that allows a user to introduce user multicast content, allowing them to pay for individual transmissions, perhaps a few hours with live video or tele-education for remote classes, workshops, etc., without the complexity of negotiating long-term service agreements.

These features and the anticipated business models are relevant to rural applications where capacity is often limited. However, the outcomes are also expected to be relevant to all next generation access networks where there is a need to pay for and deliver applications that demand high-speed transmission. Although not an issue for most applications, the hybrid satellite and terrestrial network infrastructure used in this testbed will provide an opportunity to experiment with (and overcome) any impact of transmission delay upon digital applications.

In terms of economic and commercial benefits, the testbed provides a unique, low-cost opportunity for businesses to experiment (in a safe and collaborative environment) with satellite broadband, and its associated efficiencies (e.g. Multicast cost-savings) and unsurpassed coverage opportunities (e.g. Greater potential market with higher revenues), to explore the potential impact on service delivery and business plans.

In terms of social benefits, rural communities stand to gain the most from many new servicesenabled by advanced broadband networks, including improved access to health, education and government services by dissolving barriers such as distance, time, geography, weather, and economics. DART will contribute to reducing the digital divide and improve access to NGA in remote areas, not just in terms of speeds, but in terms of the perceived quality of service e.g. many broadband/NGA services provide social benefits by delivering centralised services at distance, especially to remote areas (e.g. telemedicine). Government priorities to reduce expenditure will be assisted by allowing more economic delivery of public services without exclusion. For example, the NHS could pay for temporary QoS enhancement to support an Assisted Living application, or a specific course of treatment.

As well as cost benefits, and usual environmental savings related to the efficient delivery of services to remote locations e.g. fewer face-to-face GP visits and the corresponding reductions in carbon emissions, satellite infrastructure is also intrinsically a low carbon technology, with the launch producing only one third of the CO2 of a transatlantic flight

Technology Innovation

Innovation in the project includes:

• Exploring the integration of satellite broadband with other networks, exploring the charging models required to realise economic services and help to demonstrate the viability of a market for advanced services.
• Offering the ability to vary the Quality of Service (QoS) and user bandwidth on an occasional use basis. This is not available today, but is envisaged in future networks, it benefits the application developer (and hence user) in that it can enable a wider-breadth of services, and also the content provider (for example to offer a range of qualities and pricing models).
• Broadcast / multicast in broadband networks is an established research subject, but one that has not been widely used outside academia or closed private networks). This project brings together people allowing the technology to be explored for both live services and cached delivery. The outcomes are expected to help explore new applications and to analyse the viability of approaches to this market.
• Use of caching technology for multimedia could offer immediate benefit for caching video content. Although this is widely used for webpages, there are challenges for commercial content (including guarantee of QoS, content management models, user behaviour) that require exploration that includes all relevant stake-holders (application developers, networking technologists, network service providers, content providers/owners, user communities).
  

Experiments

Experiments will be defined and demonstrated over DART. This can explore features and the anticipated business models relevant to both rural applications as well as more widely in the Future Internet (FI), particularly with regard to video caching at the edge of networks and hybridisation of different access technologies.

The testbed will be made available at three levels:

(1) Organisations within the consortium wishing to perform their own experiments.

(2) Approved applications and content providers outside the consortium that have expressed an interest to experiment with the testbed, currently including a telehealth provider, a mobile network operator and a major vendor of telepresence solutions.

(3) Applications and content providers that are not yet on our radar and will be identified during the course of the project. TSB management of the overall programme is expected to help engage additional new users.

During the trial operations we will provide necessary training and technical support to ensure that experimenters are able to gain the maximum value from the testbed and understand its capabilities. We will develop a recruitment plan as part of the project and develop appropriate contractual mechanisms with both the experimenters and end users. Management and co-ordination of the experiments is also likely to be required to use the available network resources in the most effective way. The capability for scheduling of experiments and the ability to subset end users for simultaneous access may also need to be planned. We will endeavour to foster a collaborative environment, sharing lessons learned amongst testbed experiments, where commercial sensitivities allow.

Some example experiments include:

Satellite Cache: Integrating content caching as a part of a satellite terminal. An initial proof of concept could use a separate cache server, but prodcution service could utilise a cachelocated within the satelloite indoor unit. Caching is particularly important to support users demand for high bandwidth services and increasing consumption of video. Caching therefore has to be part of future networks and DART will provide extremely effective caching solutions based on satellite multicast, high speed local networks and intelligent services. TChallenges include how to populate the cache with the optimum content for a user, how to transport content, when to select multicast v unicast transmission and charging/copy protection models.

The companies wishing to use the demonstrator will provision the trials using a web portal which captures the following data about their trial:

• Service required: Bandwidth management, Quality of service, Micropayments
• End users: Number of end users, age profile and optional data points will be available including gender, income, etc
• Duration: Start & end date
• Reporting options: e-mail stats, via portal
• Additional services: Focus groups, interviews with end users

Work Plan

The project will be divided into two phases that will run sequentially.

Phase 1 (months 1 - 6)

• The project will design a testbed architecture that integrates appropriate tools for management of the testbed features and to support experimentation using the testbed. UoA and UoL will be supporting this objective.
• The project will implement and install the testbed infrastructure including necessary modifications to existing networks, procurement of hardware and preparation of relevant supporting documentation (procedures, user guides, relevant commercial agreements). UoL will be supporting this objective.
• The project will seek to recruit user communities and encourage best practice for participation in testbed experiments. UoA and UoL will be supporting this objective.
• The project will specify an engagement plan to guide experiments. This will provide a framework for the digital applications and content providers that will use the testbed. It will also identify the way in which experiment outcomes will be disseminated. UoA will be supporting this objective.

Phase 2 (months 6 - 18)

• This phase of work will support use and operation of the testbed including resolving problems (including making minor modifications).
• The project will engage applications and content providers (based on the engagement plan) and coordinating their use of the network to optimise network resources; and support knowledge management.
• UoA and UoL will assist in dissemination of the results and promoting best practice for evaluation and exploitation, based on the engagement plan developed in phase 1. This will include organisation of workshops as detailed in the Pathways to Impact statement.
• The outcomes of the project will be presented at relevant conferences and workshops (including participation by UoA and UoL).

University of Aberdeen Contact: Prof. G Fairhurst, School of Engineering.