Research On Key Techniques In Intelligent Networks

With the advances of Intelligent Networks (IN) and its service developing techniques, there are more and more, various kinds of IN services which would be provided to the users via telecommunication operators, where a challenging issue is to guarantee the quality of the services (QoS). In fact, the key point of IN applications is how to improve the QoS of the systems, including the management and providing of the IN services.This dissertation researches on two key techniques to improve the QoS of IN, i.e., the feature interactions detecting techniques and the IN loads controlling technique. According to the phases of the IN service developing procedure, the approaches to handling feature interactions can be divided into four classes, i.e. the approaches in the requirement-phase , the design-phase, the test-phase and the run-phase.So-called requirement-phase approach studies the feature interaction problems basing on the service behaviors which can be observed by the users, to develop methods of detecting feature, where the URN(User Requirement Specification) is employed as a specification of the feature interaction requirement and a model of service behaviors which can be observed by the users. According to this principle, this dissertation proposes a model of feature interaction, defines a rule of feature interaction detection, and introduces a theorem of feature interaction detection. Furthermore, a requirement-phase approach of detecting feature interaction is developed. Besides, the dissertation proposes a rule and a method of detecting feature interaction based on Scenario, which can be applied to dynamic detection of feature interaction.The design-phase approach studies the feature interaction problems basing on the behaviors of features, which can be observed by the designers. It builds up a feature model to represent the internal states of the feature and its transition with a finite-state machine and to represent the intrinsic properties of the features with temporal logic. Basing on such a model, this dissertation puts forward a principle to combine features and a feature interaction model. After proving a theorem of detecting feature interactions, the dissertation proposes a practical method to detect feature interactions in design-phase. At the same time, the dissertation presents a method to avoid the state-explosion in detecting algorithms.The test-phase approach studies the feature interaction problems basing on the testing viewpoint. After discussing single feature testing techniques, this dissertation proposes a method of multi-feature interaction detection, and develops an interaction detection-oriented algorithm to generate testing sequence, i.e. the test sequence generation algorithm based on UIO overlapping loop.The run-phase approach studies how to process feature interaction problems in the real running environment. It sets up feature interaction processing methods basing on the events (messages) tree in the running environment, which includes constructing trace space, cutting trace space and extracting resolution. This dissertation proposes a learning mechanism to improve the acquired resolution.Another key technique to affect IN QoS is IN load controlling. Since the traditional IN service controlling is a centralized one (single service control point), the service control point (SCP) is becoming a bottleneck to limit IN applications with increment of the service amount. Therefore, the dissertation proposes a multi-SCPs IN model and its corresponding load controlling algorithm.The proposed model composes the strongpoints of the centralized IN and distributed IN, and the corresponding load controlling algorithm contains overload control and route control, which come from the analyses of the queuing model, the call model, and the call processing procedure in the multi-SCPs IN. Besides, the dissertation proposes a load controlling method for IN based on load information feedback and criterion of the least service running cost hi SCP, moreover, analyses the fairness of th...