Note: for more detailed info, go to MEAG MIDAS page

Middleware Platform for Developing and Deploying Advanced Mobile Services (MIDAS)

Sixth Framework Programme Project

Project summary:
Project type: Sixth Framework Programme Project
Project start: January 2006
Project duration: 30 months
Project budget: 4,9M EUR

Brief project description

The project will define and implement a platform to simplify and speed up the task of developing and deploying mobile applications and services. It focuses in particular on making it feasible to provide mobile services where: (1) The number of users is very large; (2) The network may need to be set up at short notice, or for limited duration; (3) Infrastructure is limited and some users may have to use ad-hoc communications.

The platform will consist of an overall architecture, and middleware building blocks providing solutions to technical issues that must be addressed in developing mobile services.

The functions provided by the building blocks will include: (1) Maintaining connectivity between nodes, mixing infrastructure-based and infrastructure-free communications as needed; (2) Information sharing through development of a flexible distributed data management system; (3) Gathering, transforming and distributing context information in a form suitable for realizing mobile services.

A "horizontal" architecture will be used, meaning that applications can adapt to different underlying communications technologies.

The project will promote standardisation by aligning its work with the OMA (Open Mobile Alliance).

Two challenging application scenarios will be used as a source of requirements on mobility, and as a means to validate the approach: mobile support for emergency crews responding to a major incident, and a system for collecting and spreading information at sports events.

The consortium includes a major research institute [SINTEF], one of Europe's foremost system integrator companies [Capgemini], a major telecoms company [Telefonica], an SME (and IST prize winner) specialising in mobile services [Appear Networks], two organisations with expertise in the domains to be used for application scenarios [51pegasi and RATP], and two Universities with leading expertise in the technical areas addressed by the project [University of Oslo and Warsaw University of Technology].

The project duration will be 30 months, with a cost budget of 4.9 MEUR and requested grant of 2.9 MEUR. The project is expected to start in early 2006.

Project Objectives

MIDAS will develop a platform that makes it commercially feasible for the wider IT industry (not just telecom companies) to develop distributed mobile applications offering innovative services to end users. The project will focus in particular on the challenge of making it possible to develop services for events or situations where:

• There is a very large number of users (e.g. at a major sports event).
• The service has to be set up at very short notice (e.g. for teams responding to an emergency), or for a limited time period (e.g. sports event, trade fair).
• The availability of infrastructure is limited (or too expensive), and some users may in some cases have to use ad-hoc, peer-to-peer communications links.

There is potentially a large market for such services, and considerable scope for innovation. But technical difficulties and the resulting high development costs today make it commercially unattractive to develop them. To overcome these difficulties, the project will:

• Identify commonly occurring technical issues that need to be solved each time a new mobile service is designed.
• Specify and implement middleware "building blocks" which address these issues.
• Include user terminals in the service provision architecture, so that the terminals take part in providing services - not just in accessing them.
• Deal with the challenging problem of combining infrastructure-based communications (e.g. GPRS, UTMS, fixed-internet, ...) and infrastructure-less communications (e.g. Bluetooth, ad-hoc WLAN, ...) within one overall communications framework.
• Provide structured mechanisms for representing and adapting to changes in user context in a distributed, mobile environment.
• Define a service lifecycle model which takes full account of the different conditions and connectivity opportunities that may apply at different phases in providing a service.
• Develop two proof-of-concept applications to drive the work of the project, provide concrete examples of technical challenges to be addressed, and validate project results.
• Promote standardisation by aligning the MIDAS approach with the work of the OMA (Open Mobile Alliance).

The proof-of-concept applications for the two scenarios will be used as a means to measure the usefulness of the middleware platform. The criterion for success is that middleware makes it easy to implement the mobility-related aspects of the applications, where "easy" is interpreted as meaning that the time required will be in the order of weeks.

The proof-of-concept applications will use at least three different Radio Access Technologies (RATs), at least one of which will be used in peer-to-peer mode. The precise technologies to be used will be selected based on an assessment of the application requirements and the latest available technologies. Likely candidates are GPRS, WLAN and Bluetooth.

Technical Approach

The MIDAS technical approach is based on the following key principles:

Services are realized by distributed software components installed on nodes owned by the service provider and on devices operated by the end-users. When providing services for a specific event (e.g. sports, major conference, emergency situation), the service provider sets up "Instant Infrastructure": a collection of nodes and communications facilities to support service provision. A flexible and adaptable node configuration is used, customised to the conditions applying at a particular event. Nodes exchange short messages asynchronously with each other in order to maintain a Distributed Data Management System - a mechanism for sharing information. These inter-node messages are exchanged using one or more communications mechanisms - depending on what other nodes are present and what communications means are available. Application developers do not need to access low-level functionalities provided in specific mobile networks, but rather use such infrastructure as a means to transport data between nodes. The type of information exchanged can be text, sound, still pictures, video clips or measurement data provided by sensors. Middleware provided by MIDAS adapts to changes in network topology so that the application developer need not be concerned with changes in the network. This is not only to compensate for problems (such as failure of particular links or network partitioning) but also to exploit the opportunities offered (e.g. when high-bandwidth connections to central machines are possible). The overall approach to service design takes account of the different infrastructure options that are likely to be available during different phases of providing a service. As an example, consider a service for emergency crews. During the operational phase (i.e. at the scene of the accident), infrastructure may be very limited and the terminals of staff in the field will probably have to depend on ad-hoc communications. However, in the pre-operational phase - when staff are still at their base - broadband internet connectivity will probably be available, allowing downloading of security certificates, local maps etc. immediately before attending the event.

MIDAS Partners

• Stiftelsen for industriell og teknisk forskning ved NTH (SINTEF), Norway
• Appear Networks AB, Sweden
• Telefonica, Spain
• Capgemini BV, Netherlands
• Paris Local Transport Authority, France
• 51Pegasi, France
• Universitetet i Oslo, Norway
• Warsaw University of Technology (WUT), Poland