Finn Arve Aagesen

Intelligence in Communication Systems

This paper reports the findings of recent studies on the impact of intelligence in modern/future-oriented telecom networks. The ICT/ telecommunication market evolution and future trends are presented. Subsequently, the business opportunities are also discussed. The main drivers for the evolution as well as market success are mobility, security, and intelligence in the network. Though the global economy is still far from recovery, analysts expect the world telecom market in 2004 to exceed 1,000 billion Euros, with an impressive annual growth rate of between 3% and 4% Current security threats and security fiascos reveal the fact that, in spite all innovative and advanced security solutions offered by renowned companies, the security market is still quite promising. There are big business opportunities for mature solutions involving intrusion detection systems, fire walls, antivirus systems, encryption and secure mobility. The evolution towards mobility is currently passing from third-generation to fourth-generation, milli-wave lAN and HAPS solutions in the next decade. Considering the technological evolution and the current trends, the telecommunication market will present excellent opportunities for intelligent solutions and products that provide maximum bandwidth with optimal mobility, convergence between mobile and fixed networks, connectivity on demand, and finally build, transfer, optimize, manage and operate the network.

Towards a Plug and Play Architecture for Telecommunications

This paper presents an architecture specified within the project “Plug-and-play for Network and Teleservice Components” supported by The Norwegian Research Council. The hardware and software parts, as well as complete network elements that constitute a communication system, shall have the ability to configure themselves when installed into a network and then to provide services according to their own capabilities, the service repertoire and the operating policies of the system.

An approach to integrated semantic service discovery

In a distributed service environment, service discovery is a core functionality to locate the desired services. We propose an integrated semantic service discovery approach based on ontology, which provides matching of functional and non-functional properties. Functional properties are described in terms of operations, inputs, outputs, preconditions and effects, while non-functional properties are specified as business policies, QoS properties and context policies. Ontological inference and rule-based reasoning are applied for automatic and accurate discovery.

A framework for dynamic service composition

To be able to utilize the generative potential of future networks for service composition, the attributes of services and networks must be appropriately formalized, stored and made available. Important attributes are the capability and the status. Capability is an inherent property of a node or a user, which defines the ability to do something. The capability in a network node is a feature available to implement services. The capability of a user is a feature that makes the user capable of using services. Status is a measure for the situation in a system. The paper proposes a representation framework for capability and status, denoted as unified capability and status representation framework (UniCS). This framework is used to decide upon the dynamic use of capabilities, and is used to support the dynamic composition of a service system. UniCS consists of facts and configuration rules. The facts describe the availability and requirement of capabilities and status of a service system. The configuration rules verify, manipulate, transform and discover new facts with defined axioms and constraints. An instance of UniCS is the input specification for a reasoning engine to generate dynamically a composition plan for a service system.

Home energy management system for electricity cost savings and comfort preservation

The goal of the presented home energy management systems (HEMS) is to reduce electricity costs under a controlled reduction of comfort. Comfort covers here both the satisfaction of environmental states of rooms and the operating states of devices. Degree of Discomfort is a proposed measure for the dissatisfaction of comfort. A prototype HEMS was implemented. Simulations were conducted to study the potential cost savings from adjusting space heaters as well as the resulting discomfort. The cost reduction depends on the price condition and the consumers willingness to sacrifice comfort. In the extreme cases considered, the cost reduction can go up to 19%.

Capability Ontology in Adaptable Service System Framework

This paper presents a Capability Ontology (CapOnt) and a rule-based reasoning mechanism, which support service management within adaptable service systems. The ontology concepts comprise capability types, capability parameters and service management functions related to capabilities. Capability parameter values can be defined by constraints on other capability parameters. The service management functions included in the ontology are Capability Administration, Capability Configuration, Capability Allocation and Capability Performance Diagnosis. The service management functions are defined by rules consisting of constraints and management actions. The ontology concepts are represented in OWL (Web Ontology Language) and OWL/XDD (XML Declarative Description Language) – a rule-oriented knowledge representation. An intelligent conference room example with simulation results, that demonstrates the CapOnt and the rule-based reasoning, is also presented.

An Adaptable Capability Monitoring System

SNMP-based monitoring systems are founded on a platform with precompiled agents in the managed node. Software updates and re-configuration are not flexibly handled. This paper presents a solution for monitoring by use of a service framework for adaptable service systems in cooperation with traditional SNMP agents. A two-level manager system consisting of Main monitoring managers and Intermediate monitoring managers executes on TAPAS (Telematics Architecture for Play-based Adaptable Systems) platform for adaptable service systems. The Intermediate monitoring managers are federated managers that communicate with ordinary precompiled SNMP agents by using SNMP protocol. The manager part of the monitoring system will have flexibility both with respect to software updates and re-configuration. The managers are TAPAS role figures. Role figures can be deployed, instantiated and moved during run-time. This feature can be used in overload and fault situations. Some experimental results are presented for cases where the Main monitoring manager is autonomously moved after failure.

Mobility Management in a Plug and Play Architecture

Mobility Management functionality in the Plug-and-Play (PaP) architecture is presented comprising different approaches for Actor, Terminal, User and Session Mobility Management. We explore a few issues related to implementation design and propose a set of components to facilitate the deployment of this platform in the available PaP applications. The PaP architecture is briefly introduced and subsequent definitions and terms are dealt with. The architecture itself is based on a theatre metaphor, in which plays define the functionality of the system. PaP components are realised by actors playing roles defined by manuscripts. An actor’s capabilities define his possibilities for playing various roles. The mobility management is introduced to add a capability to handle any move of any constituent of the system. This could range from moving functionality, or actors, to moving users of the system. Throughout the paper we try to give a survey on the various mobility cases and make an effort to demonstrate an early set of mobility management algorithms or methods.

Intelligence in Networks

This paper proposes the use of active networking technology for congestion control and fast error correction with application level error recovery code deployed in active nodes. Its focus lies on real-time video transmission of sender based adaptive applications which are supported by active network nodes. Experiments with a prototype model show a considerable improvement with respect to video quality, error correction time, and bandwidth utilisation

A dynamic configuration architecture

Network-based services have, for more than a decade been, been an important research topic. A demand has arisen for a platform with functionalities beyond existing solutions. The paper develops a formal framework for dynamic configuration and reconfiguration of services in TAPAS - telematics architecture for plug-and-play systems (see http://tapas.item.ntnu.no). It provides representation, computation and reasoning mechanisms for semantic description and matching of required and offered capabilities and status which are required by a particular service system. It employs CIM and recently developed languages for the semantic Web in order to provide a mechanism for human-readable and machine-comprehensible descriptions of status, capabilities, system (re)configuration plans as well as exchanging messages. It also exploits XML declarative description (XDD) theory to unify such various languages seamlessly into a single uniform formalism. It permits formal definitions of application-specific configuration requirements and constraints as well as reconfiguration policies. Reasoning about these definitions and the available capabilities/status of nodes in the system yields appropriate (re)configuration plans for the composition of new services and for adaptation of current services.