Research And Experimental Verification Of Routing Algorithm In Cognitive Networks

Limited by current data networking technology, networks often lack of ability to adapt, resulting in sub-optimal performance. Limited in state, scope and response mechanisms, network elements (consisting of nodes, protocol layers, policies and behaviors) are unable to make intelligent adaptations. The interaction of information is blocked by the layered protocol architecture, making that network elements unaware of each other's status. For a network, any response to network stimuli can only be made inside of its limited scope and thee adaptations they perform are typically reactive, react after a problem has occurred. The concept of cognitive network was put forward in this background. Some of the nodes in the cognitive network have the ability of learning and reasoning. Through observing and inspecting the network environment, they can dynamically reconfigure their parameter and behaviors to adapt to the network environment and optimize the end-to-end performance.This paper studies the principle of cognitive networks, the routing technology in cognitive networks and the design of a cognitive network experiment platform. Specific tasks are as follows:1. Analyzed the current research status and development trend of cognitive network. Compared the relationship and difference between cognitive network and cognitive radio. Studied the basic concepts, framework and key technologies of cognitive network.2. Reviewed the routing technology in traditional communication networks, and discussed the routing technology for cognitive network.3. Introduced a routing algorithm framework and a simple routing example. Specially studied the multi-path routing algorithm and presented a cognitive multi-path routing algorithm for streaming media transmission. In a bandwidth-limited network, the algorithm can select effective routs that meet the video transmission's bandwidth and guarantee low delay.4. Designed and implemented a demonstration/verification experiment system which can demonstrate the network scenarios and the function of algorithms. The results show that the presented cognitive multi-path routing algorithm works well.