1 Project Summary *

2 Project Objectives *

3 Participant List *

4 Contribution to Programme/Key Action Objectives *

5 Innovation *

6 Community Added Value and Contribution to EU Policies *

7 Contribution to Community Social Objectives *

8 Economic Development and S&T Prospects *

9 Workplan *

10 Appendix A: Consortium Description *

11 Appendix B: Contract Preparation Forms *

2. Project Summary

1. Objectives

The objective of this project is to design, implement and trial a next-generation system that will enable Internet resource management through market forces, specifically by enabling differential charging for multiple levels of service.

Offering this capability will increase the value of Internet services to the customers through greater choice over price and quality, and reduced congestion.

For the network provider, flexibility will be improved, management complexity reduced and hence revenues will increase.

The above price-based resource management pushes intelligence and hence complexity to the edges of the network, ensuring the same scalability and simplicity of the current Internet.

2. Description of the Work

1. Milestones and Expected results

2. Project Objectives

2. Participant List

 

Participant Role	Participant number	Participant name	Participant short name	Country	Status*	Date enter project	Date exit project
S	P1	Hewlett-Packard European Laboratories	HPLB	UK	C-F	Begin	End
U	P2	British Telecommunications plc	BT	UK	C-S	Begin	End
S	P3	Research Centre of the Athens University of Economics and Business	AUEB	GR	P	Begin	End
S	P4	Eidgenössische Technische Hoschshule Zürich	ETHZ	CH	P	Begin	End
S	P5	Darmstadt University of Technology	TUD	D	P	Begin	End
U	P6	Telenor	TN	N	P	Begin	End

 

*C = Co-ordinator (or use C-F and C-S if financial and scientific co-ordinator roles are separate)

P - Principal contractor

A - Assistant contractor

 

4. Contribution to Programme/Key Action Objectives

1. Essential Technologies and Infrastructures

Some analysts believe bandwidth will be super-abundant and virtually free while others predict advances in computing will continue to eat up new bandwidth as it comes on line. Flat rate subscription billing is used by many ISPs but relies on significant under-utilisation of the resources subscribed to (typically under 1%). This encourages customers to waste resources paid for but not used, with a danger that the release of new robotic applications could quickly cause congestion collapse, especially for scarce resources such as radio spectrum for nomadic services. Also, low utilisation forces providers to over-book resources in order to remain competitive, eroding any service guarantees needed for broadband quality services. Connect-time billing improves matters, but is undesirable and infeasible as customers move to permanent Internet connection with the introduction of digital subscriber line access - "Internet Dial Tone".

Whether or not basic bandwidth turns out to be super-abundant, the resources needed for applications requiring low latency, loss or jitter or guaranteed bandwidth will always be at a premium. A sustainable ISP market can only exist if usage charging schemes are created so that price matches value to cost without complexity for all these services. In support of Key Action Line IV.2.4, we will define, build and validate open interfaces and standards for price communication and account reporting to enable the market managed multi-service Internet. A market mechanism is the most natural approach to allow interworking between networks, service platforms and customer systems.

2. New Methods of Work and Electronic Commerce

Electronic Commerce is expanding to include the sale of communication services that need the network to provide a certain quality of service, such as video conferencing over IP, Internet TV and interactive networked simulations for training. Senders or receivers of such data will want to have the flexibility to apportion the costs of these transfers between themselves in arbitrary combinations. Communications service providers (e.g. video streaming servers) will also want to bundle network charges for both ends with their charging on the basis of usage.

Such developments will require network charging systems to emulate e-commerce systems. We plan to enhance the network industry's customer-supplier relationships as desired under Key Action Line II.3.3. The architecture described later puts the customer's system on an equal footing to the provider's, including real-time feedback, validation and personalisation of charging, empowering the customer through a shift of control and choice over price, quality and admission control policy.

3. Systems and Services for the Citizen

Usage-based pricing system for the Internet will result in Internet facilities being more widely available. Currently, Internet access is only available to those willing to pay the fixed rate asked ("free" Internet access has costs), barring it from those who do not use it enough to warrant this payment, or cannot afford it. Enabling users to pay only for what they use will allow more people to access the Internet, low low-bandwidth users to pay less, and will also allow high bandwidth users to secure a better quality of service.

 

6. Innovation

1. Technical and Business Innovation

1. State of the Art

• D.J. Songhurst (1999), editor. Charging Communication Networks: from Theory to Practice. Elsevier, Amsterdam. ISBN 0-444-50275-0.
• C. Courcoubetis, G. D. Stamoulis, C.Manolakis, and F.P.Kelly (1999) An Intelligent Agent for Optimizing QoS-for-Money in Priced ABR Connections. To appear in Telecommunication Systems, Special Issue on Network Economics.
• G. Carle, G. Fankhauser and B. Stiller (1998) Charging for ATM-based IP multicast services. Proc. 5th Networld+Interop Conf, Las Vegas, 149-154.
• M. Rizzo, R. Briscoe, J. Tassel and K. Damianakis (1999)
  A dynamic pricing framework to support a scaleable, usage-based, charging model for packet-switched networks. In Proc. IWAN '99. Springer-Verlag.
• R.J. Gibbens and F.P. Kelly (1999) Resource pricing and the evolution of congestion control. Automatica 35. http://www.statslab.cam.ac.uk/~frank/evol.html
• M. Karsten, J. Schmitt, L. Wolf and R. Steinmetz (1999) Provider-oriented linear price calculation for integrated services. Proc. 7th International Workshop on Quality of Service, London. 174-183.
• Lee W. McKnight and Joseph P. Bailey (1997) editors. Internet Economics. MIT Press.
• J.K. MacKie-Mason and H.R. Varian (1994) Pricing the Internet. In B. Kahin and J. Keller, editors, Public Access to the Internet. Prentice-Hall, New Jersey.
• Odlyzko. A modest proposal for preventing Internet congestion.
  http://www.research.att.com/~amo/doc/modest.proposal.ps, 1997
• R. Edell and P. Varaiya (1999) Providing Internet Access: What we learn from INDEX (Internet Demand Experiment). http://www.path.berkeley.edu/~varaiya/papers_ps.dir/networkpaper.pdf

2. Community Added Value and Contribution to EU Policies

2. Contribution to Community Social Objectives

As noted in section 1, current (flat-rate) pricing for Internet access encourages waste and provides no incentive for competitive pricing to consumers and businesses. Internet charging systems such as we envisage creating are close to a perfect market mechanism and thus will expose the true costs of traditional telecommunications and encourage lowering of prices through competition, reducing the need for regulation of liberalised markets. This will result in a wider range of IP-based services coming within the reach of a greater number of people and organisations. The enhancement of Internet charging systems to include electronic tariffing, real-time per-session feed-back, receipts and payment flexibility will also make possible a wide range of services that hitherto lacked an effective pricing and charging mechanism.

By providing the correct economic incentives, flexible charging will encourage the introduction of high bandwidth applications and controllable service quality. For instance, the introduction of "3^rd Generation Mobile" services will require significant investment. The ability to implement realistic pricing mechanisms will help to lower barriers to entry for providers of new services.

Accounting for resource usage makes it easier to distinguish between high-demand and low-demand users. For low-demand users, charges can be expected to decrease compared to flat-rate charging, whereas high-demand users are charged appropriately for their resource consumption.

This will promote the development of new e-commerce services based on streaming real-time media as well as more traditional discrete data, resulting in new business and job creation. It should also help make the economic case for building systems which better support non-commercial uses of the Internet, such as education and research, and citizen access to state services and IP-based systems.

 

 

4. Economic Development and S&T Prospects

1. Dissemination and Standardisation

1. Standards Bodies

The main body for the development of Internet standards is the IETF. Relations with the IETF are already established with formal channels for providing feedback on Internet technology requirements. Some members of M3I already have prominent roles within the IETF (including a member of the Internet Architecture Board) and M3I, with its intended focus, will serve as an excellent point of contact for billing and charging issues.  Certain of the partners are also well placed to set de facto standards for charging system inter-operation.

Apart from the IETF, another important standards group that we will contribute to is the Internet Protocol Detail Record initiative. This group is working on standardising the format of the data records generated by the usage data collected and correlated from the ISP network. See www.ipdr.org.

1. Market Opportunity

1. Industrial Context

The main target for our work will be the ISP (Internet Service Provider). The Internet service provision industry is already huge and will explode as communication services become outsourced to all sizes of enterprise. Small, medium and large ISPs will all need charging solutions for the services they will sell. The nature and range of services will also increase in complexity, due to the flexibility that stems from the simplicity of the underlying building blocks. Robust, low-cost, secure, auditable and scalable infrastructures will be needed for a sound financial basis for selling these services.

The consortium deliberately invited a second ISP as a primary partner to ensure both a large ISP and a smaller ISP were represented and could afford access to their respective customer bases to improve the scientific validity of the experiments and trials. BT has a national presence as both a wholesale and retail ISP in the UK where subscription charging for Internet has recently been overtaken by connection time charging. BT also operates Internet services globally through a large number of joint ventures. Telenor has a strong presence as an ISP in Scandinavia where Internet/PC/ISDN penetration is exceptionally high in the home market. Each company represents very different business traditions and cultures.

2. Market Trends

1. Market Strategy

1. Exploitable Results from Project M3I

The exploitable results from Project M3I will be:

2. Individual Partner Strategies

1. Hewlett-Packard European Laboratories

Hewlett-Packard sees three major market opportunities for the results of project M3I.

Firstly, HP has a rapidly growing business in gathering information on resource usage within ISP networks. The data records generated are currently used for billing and traffic planning. M3I will quickly take this Internet service charging world forward and our Grenoble division will be developing new product portfolios for service management, charging and accounting middleware.

Our second area of interest is in the development of ISP business planning toolsets. These will use the statistical data collected from M3I's ISP testbed to enable an ISP to conduct business planning experiments. These tools will use the cost models, customer demand information and pricing sensitivities to explore the consequences of different pricing scenarios.

Finally, helping ISPs make best use of usage analysis and billing systems represents a big opportunity for HP's Telecomms consulting business in Europe.  HP sees consulting in the telecommunications industry as a major growth area and is actively seeking to recruit staff for this work.

2. BT

2. Workplan

1. General Description

1. Infrastructure for Market Management of the Multi-Service Internet

A good technical architecture highlights the main separations of concerns in a system and therefore leads to well-structured technical work-packages by minimising inter-dependencies. Preliminary work by the Proposers suggests the following technical architecture, which will be introduced to form the rationale for the choice of work-packages. However, it may be refined depending on requirements defined at the start of the project.

The diagram shows the network service, S_p, as a thick arrow being provided to an application or to middleware acting on an application's behalf. A similar diagram would apply to each customer in a communication (two or more). This allows multicast and unicast to be modelled generically - by connecting together any number of the above customer/provider relationships. Note that the direction of service is "outward" from the provider, whichever direction the data flows.

The internal structure of the provider's charging and accounting system (CA_p) is similar to that developed under the ACTS SUSIE project (AC320), but with single-customer granularity as described later. The present project proposal "zooms out"' to focus on and exploit the context in which such a charging and accounting system sits. Of primary interest is the pricing mechanism, which we plan to exploit to manage network load. This consists of the provider's price setting function (Of_p under the control of E_p), the customer's functions for calculating charges from the price and mechanisms for communicating prices between the two. The price can be calculated from the tariff, which is the most important part of the offer, Of, in Figure 1. The customer may install extra components (within AM_c) to control adaptive applications that react locally to pricing signals from Of_c under the policy control of E_c. This briefly describes the most abstract pricing mechanism distilled from a superset of the partners' earlier work.

Once the price is available to the customer device, it not only allows it to react to price signals, but also potentially to validate its own charges in real-time. Therefore we allow a replica accounting system, CA_c, under customer control for this purpose. Once we allow for the possibility of the customer calculating her own charges, we can consider scalability improvements such as the provider usually relying on the customer's calculations but occasionally auditing them by running its own CA_p on a sample basis.

• Applications and middleware will interface with the other three sub-systems both using them and being used by them. As far as possible, only adaptive applications will need to be modified to allow the load management system to control them, but the other charging subsystems should work whether or not applications are modified to use them;
• Charging and Accounting System, will allow applications to request authenticated reports of usage and charges on a per-session basis to enable bundling of network charges with those for other services. Traditional itemised usage charges will also be available, but whenever required rather than in monthly batches;
• Load Management, covers traditional admission control as well as emerging price-driven approaches. The former are tightly coupled to network protocols whereas we will define the latter to be cleanly separated. Monitoring tools will be required to acquire the data needed to set prices based on the policies in the ISP's business model. Then mechanisms will be needed to disseminate prices and finally software components will have to be added to applications so that they can react appropriately to the pricing signals, under the control of customer policies;
• Network Infrastructure, covers unicast and multicast forwarding as well as the main Quality of Service proposals. These are unlikely to require major changes to add the charging subsystems as measurement and monitoring can usually be added without requiring changes. However, changes to experimental congestion notification schemes may well be required to give sufficient information for charging;

1. Requirements and Validation (Work-Package 2)

This work package will specify the overall system requirements and will conduct experiments on the prototype system in order to validate it.

1. Requirements

This task will concentrate primarily on the requirements relevant to pricing and charging for communications. Requirements will not be confined to those of end-customers, whether corporate or residential; the requirements of all stakeholders will be considered. Thus, consumer protection, usability or affordability might be the primary concerns of customers, while dependability, interoperability and flexibility might be an ISP's primary concerns.

The Requirements Analysis will be a primary input to the Architecture Work-Package. It will also provide input to the ISP Business Model task of Work-Package 4 by identifying the project's assumptions concerning the direction of the communications industry. This includes such matters as possible business models and possible customer demand profiles as well as network technology assumptions and the end-to-end QoS models that are expected. Typically, multiple alternative scenarios will be allowed for in each case - charging systems should invariably not be tied to specific network technologies.

At this stage, certain assumptions define the project. We assume a multi-service network with differential charging for different levels of service. All partners are interested in how price can be used for managing the load on such a network and what the strengths and weaknesses are compared to traditional admission control approaches.

Dynamic pricing raises numerous issues of acceptability, while also potentially giving customers more choice. Customers may prefer stability of price, quality or both during a session. On the other hand, the overriding customer interest might be immediate session start despite the price being high. Traditional admission control denies the customer the utility of a session at the time it is required. This is why the project agenda focuses on the various application-specific trade-offs between price, quality and admission control, and who should make these trade-offs.

Requirements will be agreed and reported early in the project along with the agreed working assumptions. The final report of this work-package will update these initial assumptions in the light of the results of our experiments.

2. System Integration

The purpose of this task is to integrate the technology modules produced in Work-Package 5 (Pricing Mechanisms) and Work-Package 6 (Charging and Accounting System) in order to provide and support the infrastructure required to test users' responses to differential charging scenarios, and the reaction of the network to the users' responses.

3. Trials and Experiments

Small focus groups will be conducted throughout the task duration in order to gain feedback on likely reaction to specific aspects of the work. Also, basic hypotheses assumed in the modelling and requirements task will be investigated experimentally. This task aims to provide initial validation of our fundamental project motivation, that user reactions to pricing signals can be used to manage the supply of Internet services in a fashion useful to ISPs. These results will provide the justification that the commercial partners need if they are to conduct large-scale field trials of end-customer behaviour - such large-scale market trials are not part of this project.

2. Architecture (Work-Package 3)

This task will take the requirements and assumptions from WP 2 (Requirements) and the initial models from WP 4 (Economic Modelling) and produce a high-level design. This work will include technology scanning and taking decisions on the underlying technology choices for the project. This will also provide for overall technical quality control of the development activity.

This task will recommend which aspects of the system require standardisation, and which are open to competitive implementation. Such decisions are crucial to the take-up of the work in the industry. Preparation of any standards submissions forms part of the "Dissemination" Work-package.

Taking the ideas outlined in the introductory architectural description of section 9.1.1 one level further, we can further justify the choice of work-packages and their deliverables by showing that they can be used for inter-provider pricing and charging as well as for integrating with higher level commerce systems. (It must be emphasised that the Architecture Task is required for detailed validation and definition of these modules).

The architecture in Figure 1 is not just useful for representing the customer-provider relationship at the network edge. It can also represent the relationship between network providers. In this case the 'customer application', AM_c, would be just another network service in a chain ending eventually at an edge-customer, but otherwise the architecture is recursive as shown in Figure 2.



Figure 2 Inter-provider relationship as a recursive version of the architecture

The same architecture used for inter-ISP charging can represent a corporate or University network where inter-departmental charging is required.

Note that the granularity of accounting will be very coarse between providers, compared to potentially fine-grained (but still very highly aggregated) accounting for edge customer usage. Strict adherence to the edge-pricing model keeps the system scalable by isolating each customer/provider interface from the choice of granularity at other interfaces. As long as all customers are usage-charged for both sending and receiving by the network providers at all edges, both multicast and unicast charging can be achieved very scalably. Thus for real-time billing, in contrast to the SUSIE project, the charging system is confined to isolated per-customer views, rather than attempting to collate real-time data for all customers of a dynamic multicast session, for instance. Unlike in the PSTN, any re-apportionment of charges between edge-customers will be priced and cleared through an end-to-end clearing capability set up by and for the session in question rather than iteratively along the same chain of providers as the data flow. Adherence to these principles isolates core providers from the need to meter the detail of individual flows at their boundaries.

3. Economic Modelling (Work-Package 4)

This work package will provide the economic models for the Internet service market that will drive the whole project. Figure 3 shows the dependency diagram between the main economic (dark) and technical (light) parameters of the microeconomic model of an ISP. Although all parameters are intrinsically linked, we can separate into two concerns - a cost model that is primarily technical and a business model that is primarily economic. This split, shown in Figure 3 in miniature, sets the divide between two independent sub-tasks, with a third considering the wider view of the global effects of the behaviour of all ISPs.



Figure 3 ISP microeconomic models

1. ISP Cost Model

The ISP Cost Model captures the nature of the costs that underlie any Internet communication and how these costs vary with respect to the total quality demands being placed on the system. The ISP Cost Model establishes parameters which must be considered by the ISP Business model.

Initially the model will be populated with data from existing communications services, as provided by our industrial partners who are currently providing ISP services. The Trial will enable us to refine the data and our model.

2. ISP Business Model

The business model considers the likely shape of demand for network services in a multi-service environment. It will include variables for customer sensitivity to price changes, which may depend on the customer's target market segment or socio-economic criteria.

Taken together, the ISP Cost and Business Models allow prediction of the size of the future communications services market under different scenarios. The models will have lasting benefit as new scenarios arise to exercise them. It should then be possible to predict the behaviour of the Internet as a total system, at least at a high level.

3. Market Models

1. Pricing Mechanisms (Work-Package 5)

This work package is concerned with developing the technology needed to implement and manage multi-service pricing. It is also important that we also show how pricing mechanisms can be built into the existing Internet as an addition rather than requiring any significant change in the underlying technology.

1. Price Setting

This task is concerned with constructing usage based prices. The prices will be derived from the ISP business model and may change according to changes in demand (as measured in the network) or in reaction to congestion (again as detected in the network). Thus prices reflect customer value and convey marginal costs of real-time resource consumption. It must be possible to implement a range of pricing strategies, depending on the market being addressed (inter-provider, business, consumer...) and the nature of the service offering.

2. Price Communication

The Price Communication task will design and build appropriate mechanisms for communicating pricing information in a generic fashion to the user network. For instance, congestion pricing schemes described in section 5.2 use explicit congestion notification marks on packets to modify the price, but these also require the unmodified price to have been previously announced by other mechanisms.

3. Price Reaction

This task will generate the mechanisms for users, including user applications and ISPs to respond to the pricing signals communicated by the previous task, thus creating an overall engineering solution to charging. The response to price changes will be under policy control, therefore, if the party paying is different from the person using the edge-device (e.g. within corporations), the correct behaviour can still be assured as long as the payer's policy constrains the user's policy. The major effort here is to produce APIs suitable for a variety of user or application situations and linking between the user/application and the Charging and Accounting System.

2. Charging and Accounting (Work-Package 6)

This task will design then build systems to measure service usage and to charge according to the prices communicated as described in WP5. The aim will be to ensure that the accounting system is as generic as possible, with the ability to 'plug in' appropriate meters that are specialised to measure different aspects of the network and of services provided over the network. Ways of hugely improving the efficiency of metering very large numbers of events (packet granularity) have already been introduced in the architectural discussions. Whether such efficiency is possible will depend on trading off requirements set in WP2. The problem of attributing usage to individual entities will be addressed. This is a particular challenge in the case of multicast. However, the problem is tractable if the inter-provider accounting principles described under section 9.1.3 are adhered to.

A common mistake when designing charging systems is to attempt to usage-charge for higher level services using the same system and configuration as has been optimised for network charging. Because higher level services are typically provided end-to-end, each type of session typically requires a specialised accounting system. The project scope does not include building higher level charging systems, but we will demonstrate that the network charging system can provide suitable per-session real-time accounting reports (receipts) and accept payments from arbitrary sources. This will test whether we can interwork with higher level charging systems developed by certain of the partners.

Security mechanisms will have to be incorporated into the system both to avoid fraud and also to address the important distinction between measuring requests for network service and measuring actual delivered service in a way that can be trusted by customers.

It should be noted that billing as such is not a focus for this project. Our goal is to produce billing information that can be used by well-established existing mechanisms.

3. Dissemination (Work-Package 7)

This work-package (which will be managed as part of the overall project management) is established to provide focus and resources for the various publicity, dissemination and standardisation activities which are necessary for a project so have successful impact.

The objectives of this work-package are to make M3I recognised as a key initiative in establishing European leadership in the development of a standard multi-service usage billing infrastructure. To achieve this, the initial goal will be to raise the visibility of the M3I project within the FV programme and to establish it as a flagship project
within the programme. Liaisons will be established with other key Fifth Framework
projects (hopefully within some sort of cluster group). In parallel, M3I will start raising its visibility within the industry at large, thus contributing to the visibility of the Fifth Framework programme itself.

This Work-package also provides the effort for preparation of any standards proposals, as recommended by the Architecture Work-package. The project's standards approach will include not only what should be standardised, but allow us to work with whatever channel or organisation is most appropriate for the topic, and to recommend (at least in outline) contents of the standards.

4. Project Management (Work-Package 1)

The Project will be managed according to established methods based on the PRINCE (Projects IN a Controlled Environment) methodology, an emerging standard. The main leadership will be split between the two main industrial partners. One (with current experience of managing Esprit projects) will provide the Project Director, working through a Management Committee for overall management and control, and the other partner (active in IP-based service provision) will provide the Project Technical Director, working through a Technical Committee to provide technical authority and direction.

We will manage the project on a cycle of deliverables in line with the six-monthly External Reviews, and also use these Reviews to gain feedback on the continuing validity of our work, in the light of the rapidly changing business and technical environment which the "Internet" represents.

A successful project will depend on an open and flexible system for communication and information interchange between all partners working towards a similar goal, as well as interested partners not directly involved in the work-package or task. We will establish a project server to facilitate this. We also intend to promote the establishment of small working groups to ensure good links between the various work-packages, or as otherwise required.

5. Assessment and Evaluation (Work-Package 8)

2. Work-package List

B1.	Workpackage list
Work-package No		Workpackage title	Lead contractor No	Person-months	Start month	End month	Phase	Deliv-erable No
1		Project Management	P1	37	0	24		6, 9, 14, 17
2		Requirements and Validation	P6	88	0	23		1, 10, 15
2.1		Requirements	P6	33	0	3		1
2.2		System Integration	P1	22	7	22		10
2.3		Run Trials and Experiments	P6	33	4	23		15
3		Architecture	P2	37	2	10		2
4		Modelling	P3	127	1	24		7, 8, 11, 16
4.1		ISP Cost Modelling	P4	42	1	10		7, 16
4.2		ISP Business Modelling	P3	42	1	12		8, 16
4.3		Market Modelling	P2	43	1	24		11, 16
5		Pricing Mechanisms	P5	74	3	21		3, 12
5.1		Price Setting	P5	32	3	21		3, 12
5.2		Price Communication	P5	29	3	19		3, 12
5.3		Price Reaction	P2	13	3	21		3, 12
6		Charging and Accounting	P4	46	3	21		4, 13
7		Dissemination	P1	22	6	24		5
8		Assessment and Evaluation	P3		0	24		as WP 1
		TOTAL		431

 

 

 

4. Work-package Descriptions

1. Work-Package 1: Project Management

Work-package number:	1 (Project Management)
Start date or starting event:	Month 0
Participant number:	P1	P2	P3	P4	P5	P6
Person-months/participant:	15	3	2	2	2	2

 

Objectives This Work-package guarantees the success of the project through efficient and effective management of the resources allocated to this project, together with the achievement of agreed objectives.

 

Description of work co-ordinate the project activities; produce the CEC reports; resolve issues between partners and with the CEC; co-ordinate information dissemination; define resource management plans; monitor and ensure adherence to plans; ensure production of deliverables; ensure inter-project communication and cooperation Timing: this work will be on-going throughout the project

 

P6: Responsible for meeting reporting and review requirements as an individual partner

Deliverables: D6: Project Progress Report 1: including Requirements and Design D9: Project Progress Report 2: including Architecture and Prototypes D14: Project Progress Report 3: including System Integration D17: Project Final Report: including Experiments and Trial

 

Milestones and expected result: External Review 1: All reporting as required - project has attained expected status with requirements and design established External Review 2: All reporting as required - project has attained expected status with technology prototypes available External Review 3: All reporting as required - project has attained expected status with prototype systems built, trials going live Final Review: All reporting as required - project has attained expected status with results of trials available

 

3. Work-Package 2: Requirements and Validation

Work-package number:	2 (Requirements and Validation)
Start date or starting event:	Month 0 for preparation of Requirements Month 4 for design of Trials Month 7 for system design
Participant number:	P1	P2	P3	P4	P5	P6
Person-months/participant:	22	21	3	12	5	22

 

Objectives This Work-package generates the requirements from the users (both providers and customers), integrates the prototype systems and runs trials and experiments to test the assumptions and evaluate the results of the project. The trials will demonstrate the technical viability of the technology developed within realistic scenarios. The trials will investigate the technical and business implications of introducing differential charging for services.

 

Description of work Task 2.1: Requirements - Generate the requirements and detailed trial and experiment scenarios Task 2.2: System Integration - Perform systems integration of the technology modules Task 2.3: Trials and Experiments - Run trials and perform experiments Timing: work on Requirements will begin immediately at the start of the project. First results by end of month 3, validated after system integration. Trials and Experiments defined during months 4-7 after availability of Requirements. They will be executed starting from availability of the Cost and Business Models (month 10) through to the end of the project. Systems integration is an iterative process in parallel with WP 4, 5 and 6. The operational system will be available from month 16, refined during the remainder of the project.

 

Partner Contributions: P1: As supplier, contribute to definition of requirements (2.1). Build and support project test-bed. Lead the System Integration activity (2.2). Support test-bed execution (2.3). P2: As service operator, contribute to the definition of requirements (2.1). Advise on implementation requirements for integration (2.2). Provide human factors expertise for design of trials (2.3). P3: Contribute to definition of requirements, on the basis of market information on users (2.1). P4: Contribute to the definition of requirements on the basis of past experience in the area of charging systems (2.1). Contribute to system integration and test, particularly concerning integration of technology modules for the cost system (2.2). Provide input from cost module design for the trials (2.3). P5: Contribute to system integration and test, particularly concerning low-level software systems developed by this partner (2.2). Support software during trials (2.3) P6: Lead WP. Contribute to the definition of requirements as a network operator and ISP, to ensure applicability and relevance of the work undertaken (2.1). Advise on implementation requirements for integration (2.2). Run the testing, validation and practical experimentation activities (2.3).

 

Deliverables D1: Requirements Specifications D10: Integrated Trial Test-bed Systems D15: Report on Trials and Experiments

 

Milestones and expected result External Review 1: Requirements and Trial Scenarios External Review 3: Prototype systems built, Trial going live Final Review: Report of Trial and Experiments

 

5. Work-Package 3 - Architecture

2. Work-Package 4: Economic Modelling

Work-package number:	4 (Economic Modelling)
Start date or starting event:	Month 1
Participant number:	P1	P2	P3	P4	P5	P6
Person-months/participant:	8	23	24	18	8	16

Objectives This work-package will provide the models that will allow decisions to be taken on pricing structures, business models, and the feedback effects on the performance of the system that will result from applying these models to the market.

 

Timing: this work will begin as soon as the project commences. Models, together with their software implementation will be available by month 12. They will be refined and populated with actual data as the Trials proceed, through to the end of the project.

Partner Contributions: P1: On the basis of experience as a current business supplier to ISPs, contribute to the costing and business models (4.1, 4.2). P2: On the basis of experience as an operator, contribute to the costing and business models (4.1, 4.2). Provide pre-existing know-how in market management of large scale distributed systems and traffic analysis for WP 4.3 and lead this sub-WP. P3: Lead WP and sub-WP 4.2. Investigate the use of various cost models for ISPs. Construct ISP infrastructure models, decide on the use of appropriate definitions for cost (4.1). Investigate the issue of producing a model that specifies the interaction between ISPs and is consistent with the business models. Consider the various markets and the possible mechanisms for transactions. Relate business models to cost models (4.2). Use mathematical tools to analyse the results of interactions between various agents (from WP 4.2). Analyse the stability issues related to prices and the time-scales of price fluctuations (4.3). P4: Lead sub-WP 4.1. Establish the cost model for ISPs on the basis of parameters for technical performance (input also from other partners) and parameters from the business model. Develop an example where data from existing communication services are utilised and the suitability of the cost model can be shown. Participate in the trials that will allow for refinement of this model, and tuning based on real-world data. P5: Contribute to ISP cost modelling (WP 4.1) with expertise on low-level system requirements for support of certain QoS characteristics and extracting and communicating price information. Check assumptions about business and market modelling against real conditions of packet transport. P6: As an ISP, contribute to development of the cost model (4.1). Analyse business models for services in multi-service networks. In parallel, investigate optimal technical solutions for the accounting of these services in the network (4.2). Contribute to the market studies for sub-WP 4.3.

 

Deliverables D7: ISP Cost Model prepared D8: ISP Business Model prepared D11: Global interaction models prepared D16: Models populated with actual data and refined as a result of Trials

 

Milestones and expected result External Review 2: ISP Cost Model and ISP Business Model created, documented External Review 3: Global Interaction Models created, documented. Final Review: Models verified by Trials

 

 

4. Work-Package 5: Pricing Mechanisms

Work-package number:	5 (Pricing Mechanisms)
Start date or starting event:	Month 3
Participant number:	P1	P2	P3	P4	P5	P6
Person-months/participant:	9	18	8	4	24	8

 

Objectives This work package will develop the technology modules that are necessary to set, communicate and react to price in order to manage load.

 

Description of work Task 5.1: Price Setting - Develop tools for setting prices including price related data gathering. Task 5.2: Price Communication - Develop mechanisms for communicating them to users (persons or applications) Task 5.3: Price Reaction - Develop API's to enable user applications to respond to pricing information Timing: this work will begin after initial Architecture decisions are taken (month 3, WP 3). It will then be an iterative development in parallel with WP 6. Initial software design will be completed by month 6, modules incorporated in the operational system by month 16, then updated as a result of the trials during the remainder of the project.

 

Deliverables D3: Design of "Pricing Mechanism" tools D12: Integrated Pricing System

 

Partner Contributions: P1: Lead the work to build tools for price setting and general business planning, together with incorporation of feedback (5.1). P2: Lead sub-WP 5.3. Provide pre-existing know-how on client-based price reaction middleware. P3: Address the open issue of converting congestion signals to actual price values (5.1). Analyse the feedback loop that prices implement, and assumptions about the utility function of users (5.2). P4: Contribute pre-existing know-how on pricing differentiated services to the mechanisms for price setting and price communication (5.2). Investigate price communication methods, according to the Internet services models considered (5.3). P5: Lead WP together with sub-WP 5.1 and 5.2. Responsible for software design and development for implementation of price setting based on current network state and based on the QoS characteristics delivered to application flows. This involves modifying the operating system kernel responsible for packet forwarding in order to tag packets and query queues, as well as modifying reservation protocols and bandwidth brokers in order to gather and communicate data. P6: Contribute practical experience to ensure the closed economic feedback loop of service deployment and charging is as efficient as possible. Contribute to development of the mechanism for conveying to the user (human or application) information relating to usage and charges (5.3).

 

Milestones and expected result External Review 1: Software design of deliverables External Review 2: Prototypes of deliverables available External Review 3: Integrated Pricing System Final Review: Pricing System verified by Trials

5. Work-Package 6: Charging and Accounting

2. Work-Package 7: Dissemination

2. Work-Package 8: Assessment and Evaluation

2. Deliverables List

Del. No.	Del. name	WP no.	Lead partici-pant	Estimate person-months	Del. type	Security*	Delivery (proj. month)
1	Requirements Specifications	2.1	TN	36	R	IST	6
2	System Architecture and Overview	3	BT	31	R	Pub	6
3	Design of "Pricing Mechanism" Tools	5	TUD	24	R	IST	6
4	Accounting System Design	6	ETHZ	12	R	IST	6
5	Public Web-Site for project established	7	HPLB	22	O	Pub	6
6	Project Progress Report 1: including Requirements and Design	1	HPLB	9	R	Int	6
7	ISP Business Model	4.2	AUEB	36	R	Pub	12
8	ISP Cost Model	4.1	ETHZ	36	R	Pub	12
9	Project Progress Report 2: including Architecture and Prototypes	1	HPLB	9	R	Int	12
10	Integrated Trial Test-bed Systems	2.2	HPLB	42	P	IST	18
11	Global Interaction Models	4.3	LR	36	R	IST	18
12	Integrated Pricing System	5	TUD	50	O	IST	18
13	Real-time Network Accounting System interworking with Session Accounting System	6	ETHZ	34	P	Int	18
14	Project Progress Report 3: including System Integration	1	HPLB	9	R	Int	18
15	Report on Trials and Experiments	2.3	TN	10	R	Pub	24
16	Economic Models populated with actual data	4	AUEB	19	O	Int	24
17	Project Final Report: including Experiments and Trial	1	HPLB	10	R	Pub	24

*Int. Internal circulation within project (and Commission Project Officer if requested)

Rest. Restricted circulation list (specify in footnote) and Commission PO only

IST Circulation within IST Programme participants

FP5 Circulation within Framework Programme participants

Pub. Public document

4. Project Planning and Timetable



 

Deliverable number n: see section 9.4. The main milestones of the project are the External Reviews, indicated here by Deliverables 6, 9, 14 and 17 which are the Progress Reports for the Reviews.

The project will develop technology using iterative processes: one bar on the chart may represent multiple rounds of prototyping and refinement in the light of the trials. For more information on the phasing and dependencies of the various work-packages see "Timing" under "Description of Work" for the work-package in section 9.3.

 

 

6. Graphical Presentation of Project Components
7. Project Management

1. General

Hewlett-Packard Ltd takes on the role of Project Co-ordinator. This work will be performed by HP’s European Projects Office, set up at the end of 1996 to co-ordinate all of Hewlett-Packard’s work on collaborative research projects in Europe. This Office is co-located with HP Labs in Bristol England.

We identify two key roles in project management: Project Director and Project Technical Director. The Project Director will be Mr Sandy Johnstone, HP’s European Projects Office Manager, who set up this office after a career in consultancy and IT management in Eastern Europe and the Netherlands. The Project Director has overall responsibility for co-ordination of the projects, and for liaison with the CEC Project Officer. The Project Technical Director acts as technical authority with overall responsibility for the technical direction of the project: this task will be undertaken by Mr Robert Briscoe of BT Research.

2. Supervisory Board, Management and Technical Committees

The partners intend to establish a management structure consisting of Supervisory Board, Management Committee (MC) and Technical Committee (TC). The responsibilities of these groups are defined below.

1. Technical Committee

The Technical Committee is the forum for discussion of technical issues and planning. It is responsible to the MC for the review and approval of all technical deliverables. In addition, the TC identifies the necessity for workshops and performs the related arrangements. The TC will be chaired by the project’s Technical Director, Mr Briscoe. The members of the TC will be the leaders of the work-packages, plus one technical representative from any partner which is not leading a work-package. Work-package leaders are responsible for delivering the results of their work-package. The TC will meet frequently, as required by the technical demands of the project.

2. Management Committee

The Management Committee is responsible for establishing, executing, and monitoring the project and for ensuring timely reporting and delivery of results. The MC approves, authorises, amends and reviews matters concerning execution of the workplan. These include external communications; intellectual property rights (IPR) matters; recommendations from the TC; budget allocation and matters of financial control. Exploitation of project results is a key responsibility of the MC. The MC is the point of escalation for any conflicts which cannot be resolved by the TC. The membership will consist of one Management Representative from each partner. The Project Director is ex officio a non-voting member of the MC. The Project Technical Director, who is also ex officio a member of the MC will double as the BT representative. The MC will elect a chairman from among its members.

Partner Management Representatives are responsible for delivering the contribution of their organisation to each work-package and task. It is agreed that partner Management Representatives can also be Work-package Leaders, in order to reduce the number of people involved in managing the project and improve communication.

The MC intend to work by consensus, failing which it will take decisions by voting, each contractor having one vote.

We expect that the MC will meet at least every three months.

3. Supervisory Board

The Supervisory Board will consist of one executive or professorial level representative from each participant. The Project Director and the Chairman of the MC are ex officio members of the Supervisory Board, which will elect a chair for each meeting.

The functions of the Supervisory Board are to ensure that the partners give continued support and adequate resourcing to the project; to sponsor and promote the results of the project; to act as a point of escalation, should serious conflicts arise during the project.

The senior managers nominated as members of the Board are:

Hewlett-Packard Laboratories: Dr David Dack, Director

BT Research: Richard Nicol, Head of Advanced Communications Research

Eidgenössische Technische Hochschule Zürich: Bernhard Plattner, Professor of Computer Engineering

Darmstadt University of Technology: Ralf Steinmetz, Professor of Industrial Process and System Communciations

Telenor: Rolf-Bjorn Haugen, Director of Research

Athens University of Economics and Business: Prof George Venieris, Vice-Rector

The members of the Supervisory Board will be copied on Project Reports. Meetings of the Supervisory Board will be called by the Management Committee as a whole for reasonable cause, or by any one of the Partners through the Supervisory Board member for that Partner, in the event of escalation of an issue which has not been resolved by the Management Committee to the satisfaction of that Partner.

3. Project Reviews

1. External Reviews

The partners are very conscious that the "Internet" is a fast-moving concept, both technically and in business terms. Accordingly, we have structured the project to ensure regular completion of deliverables on a cycle in line with the six-monthly External Reviews. We will use the Reviews not only as a review of current progress, but as an opportunity to verify the continuing business and technical validity of the project, and to agree revisions to the work-plan should that be necessary.

2. Advisory Panel

The partners have found on past projects that it is valuable to have a panel of external referees with whom we can discuss the on-going technical and business relevance of the project. We intend to establish such a team, with world-wide participation, and are in discussion with several groups in Europe, in Canada (Lorne Mason of INRS) and Professor Pravin Varaiya of University of California, Berkeley who has already agreed to help us.

We will use this team to examine the results of the Assessment Work-package to enable us to monitor progress towards our project goals.

4. Communications and Information Flow

The participants intend to establish a secure project server with work-group capability as the central repository and access point for all internal project related information. A secure system running AltaVista Forum is available for the use of the project. Routine communications will be by e-mail. Between project meetings, we expect to organise regular tele-conferences. It is the Partners' experience that to maintain project momentum it is useful to hold a tele-conference with all partners every three or four weeks (naturally in addition to bi-lateral contacts).

External communications will be via a public web-site to be set up and maintained under the "Dissemination" Work-Package. A web server is available for the use of the project.

5. Quality Control

The project will be run under a Quality Management System successfully used in other EU collaborative projects (most recently Project "MultiPLECX" EP 26810 where Mr Johnstone also acts as Project Director). This QMS is based on the "PRINCE" (PRojects IN a Controlled Environment") methodology, which is approved by the UK Government for all IT related projects.

Technical Quality Control will be monitored as part of the "Architecture" Work-Package.

6. Consortium Agreement

2. Appendix A: Consortium Description

1. Overview of the Consortium

Participant	Role in the Project
Hewlett-Packard Ltd, Bristol UK Supplier of computer systems to the Telecomms industry.	Project Coordinator Overall responsible for running the project. Responsible for Prototype System Integration.
BT, GB Telecomms Operator and Internet Service Provider	Project Technical Authority Overall responsible for System Technical Architecture. Responsible for "Price Reaction" - linking charging to User/Applications
Athens University of Economics and Business, GR Business University with expertise in business modeling.	Overall responsible for Modeling, with specific responsibility for ISP Business modeling.
Eidgenössische Technische Hochschule, Zürich CH Technical University with expertise in architectures for multimedia communications, including charging.	Overall responsible for Charging and Accounting System. Responsible for ISP Cost Modeling
Darmstadt University of Technology DE Technical University with expertise in Quality of Service in distributed systems.	Overall responsible for Pricing Mechanisms. Responsible for network layer technology and its inter-operation with charging systems to support sophisticated business models
Telenor, Oslo NO Telecomms operator and Internet Service Provider	Overall responsible for "Requirements and Validation", with specific responsibility for Requirements and for "Trials and Experiments".

 

3. Individual Partner Descriptions

1. Hewlett-Packard Ltd (HP European Laboratories), Bristol UK (Coordinator)

1. Organisation Description

The Hewlett-Packard Company (HP) is a leading global manufacturer of computing, Internet and Intranet solutions, services, communications products and measurement solutions, all of which are recognised for excellence in quality and support.

HP has been operating in Europe since 1959 with the opening of its first manufacturing facility outside Palo Alto, California, at Böblingen near Stuttgart, Germany. From the outset, HP aimed to be a European citizen and have a balanced presence. The company has deliberately sought to grow, not only by relying on its marketing, but by actively investing in European manufacturing and R&D operations. For fiscal year 1997, HP reported a European turnover of nearly Euro 15 billion, 33% of world-wide sales. After 40 years in Europe, HP employs close to 25,000 people in fields ranging from R&D, manufacturing, sales and support. HP has substantial manufacturing operations in France, Germany, Ireland, Italy, Spain, the Netherlands and the United Kingdom - altogether 18 manufacturing operations in 10 sites. They are world-wide centres for client/server computer systems, workstations, personal computers and peripherals, test and measurement equipment and medical and analytical products and systems.

HP Laboratories is Hewlett-Packard's central research organisation and its innovation engine. Its dual missions are to help HP remain successful in current businesses while creating new business opportunities through technology innovation. Hewlett-Packard European Laboratories were established in Bristol, England in 1984, and currently employ around 200 researchers, with research projects focussed on Personal Appliances, Communications and Platforms for E-Business.

HP Laboratories has made many contributions in the areas of communications and e-commerce: current work in these fields includes participation in the TEN-Telecom Project "TTT-NET" (TEN 45615) on the feasibility of implementation of new Voice-over-IP services, and "MultiPLECX" (EP26810) on secure complex transactions.

2. Participant CVs

Sandy JOHNSTONE BSc: currently Manager of HP's European Projects Office with overall responsibility for all of HP's participation in EU Framework research. Mr Johnstone has held this position since November 1996. Previous posts include: Manager responsible for Training and Development of HP's Computer Consultancy Organisation in Europe (1986-1996); Information Systems Manager HP Northern Europe (1984-1986, based in Amsterdam); Customer Support Manager for Test and Measurement Products, HP Northern Europe (1981-1984); Customer Support manager for Computer Products, HP East/Central Europe (1976-1981, based in Vienna). Mr Johnstone is currently Project Director for Project MultiPLECX (EP26810) and was previously chairman of the Executive Committee of Project E2S (EP20563) on secure internet transactions.

Dr Huw Oliver: studied Mathematics at Cambridge University (1977-1980) where he received the MA degree. He received his MSc (1985) and PhD (1988) in Computer Science at the University College of Wales, Aberystwyth. He was Research Associate at University College Wales during 1988 and joined Hewlett-Packard Laboratories, Bristol in 1989 to work on Software Development Environments. In 1992 he moved to HP's Software Engineering Systems, Colorado turning the software development technology into product. In 1993 he returned to HP Labs, Bristol as Senior Member of Technical Staff and worked on real-time fault tolerant telecommunication systems. From 1996 to 1998 he worked on the ACTS "ReTINA" (Real-Time TINA-compliant Distributed Processing Environment) project including an appointment as Senior Visiting Research Fellow at Lancaster University. He has worked as Technical Reviewer of a European ESPRIT project and is currently Manager of Hewlett-Packard's Internet Research Institute.

3. BT, Ipswich GB

1. Organisation Description

BT is an international telecommunications and Information Technology group with a turnover of about £18.2 billion in 1998/9. BT is one of the world's leading suppliers of fixed and mobile communications services. In the UK, we support around 27 million customers' lines and, through our 60 per cent stake in BT Cellnet, over three million mobile connections. Our main services are providing applications, services and networks on a local, national, and international basis for residential and business users in the areas of voice, data and multimedia, through fixed, mobile and Internet communications. For further details see http://www.bt.com

BT R&D: BT's main research site occupies 40 hectares at Martlesham Heath in Suffolk, some 70 miles north of London. Approximately 4,000 people are based there; including 700 people dedicated to researching future technologies, systems, networks and services.

BT spent £268 million on R&D in 1998/99 this includes £54 million invested in longer-term corporate research. Of this around 75% stays with our in-house researchers, the remainder being invested in pre-competitive, university and other collaborative research initiatives including European funded programmes. BT is an active participant in European collaborative R&D both through its shareholding in EURESCOM GmbH and through its participation in EU Framework Programmes. BT has participated in all the EU Framework Programmes, mainly in the telecommunications and IT Specific Programmes. In the 4th Framework Programme, BT participated in a total of 23 projects in the ACTS, ESPRIT and Telecommunications Applications Programmes.

2. Participant CVs

Bob Briscoe, MA Cantab. In the mid-1980s Bob Briscoe managed the transition to IP of many of BT's research networks. Later, he represented BT on the HTTP working group of the IETF and other application layer working groups. He was BT's representative on the ANSA consortium, a successful industrial consortium that grew from an early ACTS project and resulted in the creation of the OMG and CORBA. His published research background is in platforms for global Internet commerce and the integration of Web and distributed object technologies. Since early 1997, he has led BT's team researching charging for Internet services. This covers charging for differentiated network transmission and for real-time Internet content sessions, especially over multicast. He has filed six patents in this field and published papers on scalable charging systems; managing network loading using charging; future business models suitable for connectionless networks; and scalable key management solutions for multicast. He is also studying for a part-time PhD on Internet charging at University College London.

Mary Jones MSc is a Technical Group Leader in BT's Cognition and Perception Laboratory at BT Labs in Ipswich UK. Mary has over 10 years experience in a variety of technical roles including the design and running of formal psychology experiments and the design and evaluation of customer products including physical products such as payphones, speech dialogues and graphical user interfaces. This includes several years experience in assessing the usability and acceptability of web sites.
Mary's current team consists of psychologists, social researchers and designers conducting qualitative and quantitative social research into consumer lifestyles. The Cognition and Perception lab contains people experienced in the design of formal experiments into multimodal perception and the factors influencing quality assessment. Mary is currently the BT project manager for Eurescom project P904 researching The Impact of Telework on the Quality of Life. She is also BT's practitioner representative on the British Computer Society HCI committee.

Dr Mike Rizzo studied Computing and Mathematics at the University of Malta. In 1990 he joined the Computing Laboratory at the University of Kent, where he conducted research in the areas of human computer interaction and distributed office communications systems. Upon obtaining his PhD in 1996 he returned to the University of Malta as a lecturer, during which time he was also involved in setting up the University's Computing Service and establishing the Malta Internet Foundation. In 1998 he joined the Distributed Systems Group at BT Labs, where he is currently involved in projects related to Internet charging and session control.

Jérôme Tassel, MSc, graduated from Kent University, UK. He has commercial experience in network management (for the UK National Health Service and the French Ministry of Defence) and software development for distributed systems (for the UK Ministry of Defence). He joined BT research in 1997 on a project looking at middleware to support large scale multicast multimedia communications. His current research interests lie in the engineering aspects of charging for multi-service IP, middleware for mobile devices and resource control both in the network and end points. He has published and co-published half a dozen papers in the field of network management, QoS control and charging for multi-service packet switched networks.

Konstantinos Damianakis, MSc. in Distributed Systems from Kent University, UK, studying for a PhD on Virtual Organisational Transformations. He has technological expertise in using and charging for Quality of Service protocols. He joined BT research in 1997 on a project investigating Collaborative Virtual Environments and QoS issues. His research is currently focused on building customer software and mobile agents to assist dynamically priced charging environments in multi-service IP networks. He has co-authored a number of publications in this area.




 

 

5. Athens University of Economics and Business, GR

1. Organisation Description

Athens University of Economics and Business (AUEB) is a dynamic institution of higher education in Greece, which was founded in 1920. It is divided into six departments: Economics, Business Administration, Informatics, Marketing and Operational Research, Statistics, International and European Economic Studies. Each Department offers first and Masters degree. The academic staff of the University is about 150 and the total number of students in the University approximates 5,000. The University also offers 5 MSc courses including an MBA and an MSc in Information Systems.

Research, both theoretical and applied, has always been considered as one of the University's most important tasks and is mostly implemented through the Research Centre-Athens University of Economics and Business (RCAUEB). The purpose of RC AUEB is to co-ordinate and facilitate research by members of the University’s faculty, in collaboration with other university institutions, public entities and organisations, international organisations, etc. The area of research includes Economics, Management European Studies, Finance, Accounting, Marketing, Informatics, and Statistics.

The Department of Informatics pays special attention to research, which is evident from the 15 Ph.D. students that are currently working on their Doctorates. There is also experience in international collaboration and exchange of research and teaching staff especially through the EC ERASMUS scheme, where the department has been actively involved in since1990. The Department has 4 special laboratories in the areas: Artificial Intelligence and Decision Support Systems, Data Bases and Information Systems, Networks and Distributed Systems, and Parallel Computing with an overall IT Investment of 1.5 Million ECU. The Networks and Distributed Systems laboratory is specialized in Internet and multimedia research, with emphasis in QoS provisioning and the integration with switching technologies such as ATM and Gigabit Ethernet. An important activity is targeted in pricing communication networks. Emphasis is given in pricing SLAs for Internet services, and a recent application was a project on designing the tariffs for the Greek Research and Academic Network. This is an IP network using an ATM backbone, which connects most Greek cities with large academic and research institutions.

There is also a strong collaboration with the department of accounting and finance, specially in the area of telecommunications accounting and pricing. In the past years, the faculty from both departments has collaborated in conducting studies on behalf of the Greek regulator (National Telecommunications Commission) for determining the regulatory policy in relation to interconnection pricing and unbundling of the local access. Also there has been strong involvement in studies conducted for the Greek PTT concerning cost-based pricing of its services, including Internet services.

Finally, we must stress the close synergy that exists with the E-commerce group. This is an extremely successful group that has been established since 1991. The group at the moment has 12 full-time researchers and is involved in more than 10 research projects financed by the EC or/and the Greek government, and has published a large number of scientific papers.

2. Participant CVs

Costas A Courcoubetis is Professor-designate at the Department of Informatics, Athens University of Economics and Business. He will take up this post on 1 September 1999.

He is currently Professor at the Department of Computer Science, University of Crete, Heraklion, Greece. Other activities include acting as Head of Centre for Telecommunications and Networks, University of Crete, responsible for the planning, implementation and support of the complete networking infrastructure (local networks and backbone) of the University of Crete located at Heraklion and Rethymnon. He is also Head of Telecommunications and Networks Department, Institute of Computer Science, FORTH, Greece. The activities of the Telecommunications and Networks group include Pricing of network services and Basic research in resource allocation problems for broadband networks. Previous to his appointment in Crete, Prof Courcoubetis was Member of Technical Staff, Mathematical Sciences Research Centre, AT&T Bell Laboratories, Murray Hill USA, between 1982 and 1990.

George C Polyzos is recently appointed Professor at the Department of Informatics, Athens University of Economics and Business. From 1988 to 1998 he was a faculty member, Department of Computer Science and Engineering, University of California, San Diego. In this capacity he acted as Co-director of the Computer Systems Laboratory; directed research in the areas of operating systems, communications networks, Internet protocols, wireless and mobile communications, distributed multimedia systems, and system performance evaluation. He was a member of the Steering Committee, Centre for Wireless Communications, UCSD School of Engineering. Prof Polyzos has taught graduate courses on Computer Systems Performance Evaluation, Communications Networks, Wireless Networks, and Multiple Access Communications, and undergraduate courses on Operating Systems, Computer Networks, the Internet and the World Wide Web, and Scientific Applications of Computing.

6. Eidgenössische Technische Hochschule, Zürich CH

1. Organisation Description

The department of ETHZ participating in this project is the Computer Engineering and Networks Laboratory (TIK, Institut für Technische Informatik und Kommunikationsnetze, http://www.tik.ee.ethz.ch).

The TIK is part of the Electrical Engineering Department of the ETHZ. It has carried out previous work concerning networking, multimedia, and application issues in a technology-driven approach, respectively, in various international, European, and Swiss projects. Ongoing work covers similar topics and in addition charging and accounting as well as security-related projects. The current list of projects encompass amongst others: ANN (Active Network Node), CATI (Charging and Accounting Technology for the Internet), Cobrow (Collaborative Browsing in Information Resources), VersaKey (Key Server and Multicast Security Support), SMAN (Service Management in Active Networks), and Telepoly (Tele-teaching).

In particular, end-system support and communication middleware utilising well-known network technology such as IP-based (Internet Protocol) on Ethernet and advanced high-speed technology such as ATM (Asynchronous Transfer Mode) are investigated at TIK, concerning the development of integrated service architectures. Furthermore, applications have been developed to gain practical experiences in exploiting application knowledge for the support of sufficient networking tasks. The use of advanced high-performance networking technology and the discovered scarcity of end-system and network resources, directly leads to the question of gaining a trade off of guaranteed service quality and expenses for particular applications. Commercial usage of the Internet requires a dedicated level of service guarantees and an integrated service
architecture, covering Integrated Services (IntServ) as well as Differentiated Services (DiffServ) approaches as well, which must include charging mechanisms. The answers have to be found in detailed investigations of accounting and charging of to be identified resources and services within the Internet environment.

2. Participant CVs

The project lead for TIK, ETHZ will be Dr. Burkhard Stiller. Dr. Stiller received his diploma degree in computer science and his doctoral degree from the University of Karlsruhe, Germany in October 1990 and February 1994, respectively. From January 1991 until September 1995 he has been a Research Assistant at the Institute of Telematics, University of Karlsruhe, being on leave in 1994/95 for a one-year EC Research Fellowship at the University of Cambridge, Computer Laboratory, UK. Since November 1995 he has been with the Computer Engineering and Networks Laboratory TIK, Swiss Federal Institute of Technology ETH Zürich, Switzerland as a Lecturer for multimedia communications and a Research Associate in the same area.
Aside from a number of project management tasks and participation in national research projects of Germany, Switzerland, and the UK, his primary research interests include architectures for multimedia communication subsystems, Quality-of-Service models, charging and accounting systems, resource reservation, transport protocols,
tele-teaching, and ATM networking.

 

8. Darmstadt University of Technology DE

1. Organisation Description

Darmstadt University of Technology is one of the best known universities in Germany, especially in the technical disciplines. The chair for "Industrial Process and System Communications (KOM)" at Darmstadt University of Technology, headed by Prof. Ralf Steinmetz, was founded in 1996. It is active in the communication and multimedia systems research areas. KOM cooperates closely with the GMD IPSI institute and also with various industry R&D laboratories and participates in European projects such as the Hypermedia News on Demand (HyNoDe) ESPRIT project.

The "Multimedia Networking" group within KOM, headed by Dr. Lars Wolf, is working on transmission of multimedia data via computer communication systems, Quality of Service provisioning, new communication system architectures, and especially charging for such communication services. The results of our research are published and presented in scientific journals and conferences.

2. Participant CVs

Dr. Lars Wolf (Management Contact) studied computer science at the Technical University of Braunschweig, Germany, and the University of Erlangen-Nuremberg, Germany, where he received the diploma degree in 1991. Dr Wolf received the doctoral degree from the Technical University of Chemnitz in 1995 in computer science. In 1996 he joined the Technical University of Darmstadt. There he leads a research group working on multimedia communication and quality of service in distributed multimedia systems. Dr Wolf lectures on distributed multimedia systems, quality of service, and networking in general. He is (co)author of more than 45 scientific publications and served on program committees and the editorial board of multimedia and communication systems workshops and conferences and journals. He is a member of the ACM, the GI, and the ITG.

Martin Karsten (Technical Contact) holds a diploma in business and computer science from the University of Mannheim, Germany. He worked as a student contractor on
distributed multimedia software projects at IBM's European Networking Laboratory, Heidelberg. He did his diploma thesis at University of Waterloo, Canada, creating a parallel debugger for parallel software and produced a prototype that were ready-to-use for educational purposes. In December 1996, he joined the Technical University of Darmstadt, where he is currently pursuing his PhD. Over the last years, he did intensive research in the area of charging and pricing for packet-switched network communication and published several papers and articles about this topic.

10. Telenor, NO

1. Organisation Description

TELENOR: The Telenor group (Telenor) comprises the state-owned joint stock company Telenor AS and its subsidiaries. Telenor is Norway’s major telecommunications operator with about 18,700 employees in Norway and 2,700 abroad.

Today, Telenor is the only company in Norway that offers a full range of services within telephony, data, mobile and satellite communications, Internet and IT. The company’s international strategy is focusing on the following areas: Satellite communications, directory services, Internet operations and mobile communications.

TELENOR R&D: Telenor Research and Development(R&D) is a separate unit within Telenor AS and its main objective is to lay the foundations for Telenor’s long term development, through highly qualified research within Telenor’s core areas. R&D is concentrating on long term research within three main areas: networks, services and applications and has established a multi-disciplinary programme for the next generation of the Internet which incorporates expertise in infrastructure, middleware services, and multimedia.

2. Participant CVs

Dr Ragnar Andreassen obtained his master degree in Physics at the University of Oslo in 1988, and his Ph.D in Telecommunication Engineering at the Norwegian University of Science and Technology, Trondheim, in 1997. He has since 1988 been with the research department of Telenor, working in the fields of data communication protocols, multimedia systems and network traffic engineering. He is currently responsible for Telenor research activities on Internet pricing and charging schemes. International experience includes standardisation activities in ETSI and ITU, and participation in the previous EU/ACTS project AC039 CA$hMAN. Ragnar Andreassen led the work in the field of experimentation and trials for usage based ATM charging carried out in that project.

Dr. Terje Jensen is a senior research scientist at Telenor Research and Development. His main interests include teletraffic modelling and analyses applied on various aspects of telecommunication networks and service provision. After obtaining his MEEE degree from the Norwegian Institute of Technology in 1986, he was engaged by Telenor. From 1987 until 1991 his duties covered regional administration of network planning and operation support. He received his Ph.D. degree from the Norwegian Institute of Technology in 1995. From 1994 he has been working at Telenor R&D.

Activities in recent years are related to network dimensioning, performance evaluations, quality of service framework, techno-economic analyses and technology strategy. Several of these activities are carried out as parts of international engagements, including projects within ACTS (Advanced Communications Technologies and Services) and EURESCOM (European Institute for Research and Strategic Studies). Co-ordinating roles within these projects have also been undertaken. He has taken part of the authorship of more than 50 publications and lectures for international and national conferences and journals.

 

4. Appendix B: Contract Preparation Forms