The Study And Implementation Of JadexT: A Multi-Agent System Development Platform For Telecommunication Applications

With the evolution of networks in telecommunication and the convergence of IT and telecommunication, multiple heterogeneous networks works together to provide telecommunication services. Session Initiation Protocol (SIP) plays an important role in telecommunication industry and is becoming the crucial technology of next generation network. SIP has advantage of enabling the combination of different IP-based telecommunication services and the coverage of heterogeneous networks under a single open framework, thus greatly reduces the cost of design and deployment of new services, and promotes the development of the telecommunication industries and markets.Telecommunication applications are typical distributed and usually scatteredly deployed throughout the network. Meanwhile telecommunication applications are normally open systems, and new services can be added to the system dynamically in order to meet the new requirements of users. Such complex systems are hard to model, analyze and implement. As the constantly enriched demand of users and the continued expansion of the telecommunication market, telecommunication applications become more and more complex, therefore it is a challenge for traditional software engineering technologies to develop telecommunication applications. Agent-oriented methodologies provide high-level abstraction and flexible execution mechanism and can be used to develop complex systems in a natural and effective way. Applying agent-oriented technology into telecommunication applications leads to be great research value and practical value.The thesis introduces agent-oriented software development technology as a SIP-based next generation application development paradigm to handle the complexity and interoperability of telecommunication applications. A multi-agent development platform called JadexT is developed by extending Jadex BDI reasoning engine for telecommunication applications. An example is also studied to illustrate our approach, which manifests that our approach is helpful to gain cost-effective services and develop new services rapidly and leverage applications and data across multiple heterogeneous networks.