Details of Grant 

EPSRC Reference:	GR/J48405/01
Title:	DETECTING AND AVOIDING FEATURE INTERACTIONS IN OPEN NETWORKS
Principal Investigator:	Magill, Professor E
Other Investigators:
Researcher Co-Investigators:
Project Partners:
Department:	Electronic and Electrical Engineering
Organisation:	University of Strathclyde
Scheme:	Standard Research (Pre-FEC)
Starts:	01 October 1993	Ends:	30 September 1995	Value (£):	69,104
EPSRC Research Topic Classifications:	Networks & Distributed Systems
EPSRC Industrial Sector Classifications:
Related Grants:
Panel History:
Summary on Grant Application Form
The feature interaction problem threatens the underlying philosophy of open networks - namely that services can be added rapidly and easily without threat to existing services and customers. The aim of this project has been to develop a Feature Interaction Management System (FIMS) which detects and avoids the effects of feature (or service) interactions. The FIMS must be transparent to the service creation process and thus allow the rapid provision of new services while causing minimal service delay to existing ones. Service privacy and independence between service providers must also be preserved.Progress: A new testbed has been created which simulates the Intelligent Network (IN). This was used to model the behaviour of services, their interactions and provide a platform for testing feature interaction management techniques. The ITU-T Q1200 recommendations provided the IN model to be used and six features have thus far been implemented: call forwarding unconditional, originating call screening, terminating call screening, call waiting, automatic callback and call forwarding on busy. The testbed uses a SUN workstation to simulate the operation of the IN and a PC to provide a mouse-driven graphical user interface. The workstation and PC communicate via an ethernet interface. The testbed was used as a platform for testing a technique to detect service interactions within an IN environment. This technique employs detection feature managers (D-FMs) to monitor the operation of each active service in the call. The role of each D-FM is to compare the sequence of input events and network resource state with those listed in a service signature store. Deviations from the signature sequences represent a fault in the operation of the service and thus a potential service interaction. The signature sequence stores are obtained through a learning phase during which the D-FM monitors an error-free service and saves the sequences in the signature store. A series of experiments were performed to test the effectiveness of the D-FM technique using combinations of the six services in single-subscriber and multiple-subscriber configurations. These verified the D-FM techniques ability to detect service interactions in a wide range of situations. Once it is possible to detect an interaction accurately, the information about the service and its environment at the point of interaction can be used to prevent the same interaction re-occurring. This combined system of D-FMs and feature interaction predictors will comprise the complete Feature Interaction Management System. Research has also been undertaken into artificial neural networks (ANNs) because of their pattern matching and learning abilities and their potential speed. Using the Aspirin/Migraines tool, experiments were carried out to determine the feasibility of using ANN technology in the D-FMs. However, the need for extensive training sets and long training times suggested that conventional table lookup approaches are more suitable. Research is also currently being conducted in distributed operating system error-recovery techniques and learning automata.
Key Findings
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Potential use in non-academic contexts
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Impacts
Description This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk Summary Date Materialised
Sectors submitted by the Researcher
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Project URL:
Further Information:
Organisation Website:	http://www.strath.ac.uk