Network Behavior And The Utilization Of Network Information

The network technologies trend to be intelligent and social, which makes the network environment become increasingly complex. In the self-organized network environment, it has a high complexity to get what you need from the vast ocean of information. Therefore, new theories and methods are needed for complex network environment. Firstly, we need to study the information dynamics theories in the network environment. And on this basis, we should establish the theory and methods of the information exploitation. This study not only has important scientific significance, but also will generate huge economic and social benefits.Currently, the modeling work of network behavior is still at the initial stage of focusing on interactive communication. The study of modeling intelligent behavior is deficiency. This thesis classifies social behavior and proposes behavior factors to quantify. The five elements model is proposed for predicting dynamic behaviors. Both of the quantitation of five elements model and exploitation of network information need an efficient algorithm with parallelism. This thesis uses the rules of animal group behavior. And the swarm classification mechanism and the concept of leadership are introduced to propose Group Mosquito Host-Seeking algorithm (GMHSA) and Distributed Sensor Networks Group Search Optimizer (DSNGSO). GMHSA has the advantage of simple computation unit, fast convergence and high parallelism. DSNGSO is used as the information exploitation method for sensor networks. It is used to solve single objective problem, multi-objective problem and make social intelligent decisions with the advantage of fast and global convergence. We give theoretical analysis and experimental verification for the proposed model and algorithm to its effectiveness and superiority.