Dissertation > Industrial Technology > Radio electronics, telecommunications technology > Wireless communications > Radio relay communications,microwave communications

Research of Dynamic Resource Management and Allocation Strategy in Cognitive Radio Networks

Author ZhouChao
Tutor FengWenJiang
School Chongqing University
Course Communication and Information System
Keywords Cognitive radio Resource management and allocation Non-cooperativegame EWA learning game Utility function
Type Master's thesis
Year 2012
Downloads 39
Quotes 0
Download Dissertation

As the explosive growth of wireless communications business needs and thecontinuous increase of the QoS requirements, wireless communication network shouldconstantly seek the technical approach to improve the system capacity and coveragewith the limited spectrum resources. As an intelligent spectrum sharing technology,cognitive radio has the environmental consciousness, behavior consciousness andpurpose consciousness, meanwhile,it allows multidimensional spectrum reuse andsharing in the time, frequency and spatial domain. Game theory is a mathematicalmodel to solve the decision-making problems and balance of decisions. Competing forlimited network resources between communication users can be modeled as a gamemodel, with it, an efficient and reliable resource management and allocation strategycan be designed.This paper focused on the dynamic resource management and allocation strategiesof the cognitive radio network, used non-cooperative game theory to establish themathematical model. The feasibility of solving the problem of cognitive radio networksresource allocation with game theory was verified by analyzing the Nash equilibriumand its characteristics. Then, two kinds of resource management and allocationalgorithm based on game theory were proposed:For the Centralized cognitive network, proposed joint resources evaluation andallocation algorithm based on non-cooperative game (JREA). The algorithm collectsstatistical data through the base station, carries out the quantitative evaluation of thespectrum occupied by the primary user and the scheduling management for of cognitiveusers in network, determines the maximum transmit power of each user in the networkthrough the tolerable interference threshold of the primary users, then we designed theutility function of the power allocation algorithm considered with the the QoS demandsof cognitive users and analyzed its convergence. The simulation shows that theproposed algorithm converges fast, can still ensures each user’s SINR value higher thanits QoS demand threshold in increasing noise environment, reduce the interference toprimary users, and enhance network performance.For the distributed network architecture in which users are unable to obtaincomplete information. We designed user access and signaling interaction process, builtthe priority table of available channels in network, built EWA (Experience-Weight Attraction) game learning model, and propose a channel selection learning algorithmbased on channel priority and interference between cognitive users. Simulation resultsshow that the algorithm can adaptively select channel with the best usability throughlearning of historical experience, increase the effective throughput of the system andhave a better equity in the resource allocation.

Related Dissertations
More Dissertations