Research On Information Awareness Technology Oriented To Cognitive Networks

With the diversification of user's requirement, the network systems have become extremely complicated which resulted in the unreliability of network and end-to-end performances. The present network elements could not perceive the action of others and the environment state due to the constraint of traditional network architecture. As a result, the adaptation ability of network elements for complex environment is inadequate and the quality of service can not be guaranteed. Therefore, the concept of cognitive networks is generated under this background and become an important topic for the researchers in network fields. Cognitive networks are considered as the core and development trend of the next generation network (NGN) in commom, which now are on its infant stage, the design complexity of cognitive network beyond the original understanding of system designers. By cognition of network-aware model, it completes the cognitive network-oriented information awareness, and lays foundation for the research on cognitive networkFirstly,accounting for the critical technology of cognitive network and classical network architecture from a macro point of view, a architecture of cognitive network based on MDE(Monitor-decide-Execute) loop is proposed in this paper. Based on the architecture, the cognitive layer is designed in detailed, and a sensor model for distributed cognitive network is built, where the sensor function is realized on the approach of designing the Monitoring modular of MDE. The model that is aware of the information of network is a foundational work for the research of cognitive network.Secondly,the process of information sensor is divided into three parts: information acquirement, information processing and information forecasting which are separately discussed. On the use of SNMP and NetFlow sensors, the original network information is acquired, and a hinder Markov model is presented and used to predict the state of the model. Moreover, the abnormal events is detected by the ports classified based on TCP protocols, which lead to the realization of cross-layer in the cognitive network in a indirectly manner.Finally, we designed experiment according to the information perceived from the low-and high-level process of evolution. The model of the false alarm rate and the detection rate are in accordance with the current situation of network, so, the rationality of this paper was verified.