Font Size: a A A

Research Of Cognitive Dynamic Spectrum Access In Two-Tier Heterogeneous Networks

Posted on:2015-10-29Degree:MasterType:Thesis
Country:ChinaCandidate:F Z YeFull Text:PDF
GTID:2298330467462300Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
Deployment of small cells (SCs) can provide high network capacity and extend the coverage for macrocells. However, when more small base stations are deployed or the traffic loads of macro users become heavier, interference becomes the major obstacle impacting the potential gain of SCs. In this paper, dynamic spectrum access is in depth study to alleviate interference. Three different dynamic spectrum access strategies have been proposed.The first one is the cooperative dynamic spectrum access (CDSA) in small cell networks with interference mitigation and QoS guarantee. With CDSA, a fusion centre server can formulate the optimal sensing period and perform optimal RBs allocation and power control for sinall cell network, which not only achieve interference mitigation, but also provide QoS guarantee for small users. Simulation results demonstrate that the CDSA performs much better on interference mitigation and QoS guarantee than the fixed and the randomized schemes.The second one is call uncooperative dynamic spectrum access (NCDSA) based on fuzzy logic game theory. NCDSA constructs a game model in the distributed small cell network with a utility function evaluating intra-tier interference and QoS. With the fuzzy inference system which combined chicken game theory, the cognitive small base station can perform appropriate spectrum access strategy. Simulation results demonstrate that the NCDSA outperform than the DSA using best response power control method in collocated small cell.The third one is an uncooperative dynamic spectrum access (FQUCDSA) base on fuzzy Q-learing. FQUCDSA is the improvement of UCDSA, which combined the Fuzzy inference system in UCDSA with reinforcement Q learning system. A reword function evaluating intra-tier interference and QoS has been built, and the small cells adjust their spectrum access strategy to maximize their reword. Simulation results demonstrate that FQUCDSA converges to the Nash equilibrium faster than UCDSA. Moreover, FQUCDSA has great chance to reach the optimal Nash equilibrium point uncooperatively.Thought the analytical and simulation result, a conclusion can be drawn that these three dynamic spectrum access strategies perform better than some existed spectrum access strategies in interference mitigation and QoS guarantee.
Keywords/Search Tags:cognitive dynamic spectrum access, interferencemitigation, QoS guarantee, fuzzy logic game theory, fuzzy Q-learning
PDF Full Text Request
Related items