Resource Management Mechanism For QoS In Cognitive Networks

Future network will direction towards the intelligence and adaptability. Cognitive network as the development trend of next generation networks, came into being. It can perceive current network resource information, and then based on these resource information for planning, decision-making and action. Cognitive has the ability of learning and adaptivity, so as to achieve high-level end-to-end goals of quality of service (QoS).Resource management as one of the key technologies of cognitive networks, the main research is to design a kind of adaptive, self-optimizing resource management in the cognitive network environment,so as to protect the user's end-to-end QoS .For the characteristics of cognitive network, we first give the resource management objectives and principles. Resource management research includes: the perception of resources, resource description, resource monitoring, resource allocation and scheduling of resources in five areas. This paper focuses on resource monitoring and resource allocation. With The introduction of strategies and feedback mechanism, we propose a satisfaction-based resource adaptive control framework in cognitive network (RACF_CN). In the study of adaptive genetic algorithm (AGA), we propose a resource allocation algorithm base on improved AGA (RA_IAGA), the algorithm can guarantee end-to-end QoS by dynamically adaptive scheduling system resources, so that to achieve optimal customer satisfaction. Simulation results show the correctness of the algorithm.Finally, by extending the SNMP++ protocol, we programming of the cognitive network resource monitoring system. By testing the main modules of the system , the results show that the resource management system can improve resource utilization, protection of the user's QoS.