Research On Spectrum Sensing And Network Access Techniques For Cognitive Radio Networks

The rapid increasing wireless standards and rich network applications have brought a great pressure to the limited network bandwidth,the scarcity of spectrum resources is becoming a main obstacle to constrain the development of wireless network technologies and network applications.However,limited to traditional static spectrum management,the average of proportion of the total available spectrum used by people is only 2%-6%,which causes the contradiction that for one hand the spectrum resources is very scarce,and for other hand the spectrum usage is extremely low.In order to maximize the potential usage of spectrum resources,cognitive radio is put forward.The key idea is cognitive radio has cognitive ability,and exchanges information with environments,utilizes spectrum holes according to spectrum sensing,and improves the spectrum utilization ultimately.Based on cognitive radio,researchers expand cognitive computing from radio to network domain,and introduce the concept of cognitive radio networks.Compared with existing wireless networks,cognitive radio networks can sense changes in internal and external environments,adjust the modules and configuration parameters in a real time manner,dynamic and intelligently adapt to the environment to guide future decision-making,and achieve a reasonable sharing and optimizing the useage of network resources at last.Cognitive radio networks become the main means to solve multi-network coexistence under the condition of resource-constrained,and have the ability to improve the end-to-end performance of wireless networks finally.Cognitive radio networks not only provide new model of future wireless spectrum management and operation for network managers and operators,but also provide more favorable experience to users in dynamic and heterogeneous network environments.However,the time of cognitive radio networks proposed is short,and the application environments is complexity,although basic research results have been achieved in spectrum sensing,spectrum access and sharing,there is still insufficient in sensing accuracy and access efficiency for specific network environment(dynamic and heterogeneous,etc.)and user behavior(selfish and malicious,etc),it still has large gaps to practical application.Accordingly,this paper aims at solving spectrum sensing and network access problems for cognitive radio networks under specific conditions.Based on the theory of cognitive computing,this paper introduces game theory,trust computing and fuzzy neural networks as problem-solving approaches.Combining with existing research results,this paper establishes spectrum sensing model for cognitive radio networks to meet specific network environment and user behaviors,and develops novel methods of spectrum access and network access with the ability to enhance secondary users QoS and users' experience,provides references for the promotion and improvement of cognitive radio networks technologies.The main research contents are organized as follows.First of all,a cooperative spectrum sensing method based on non-cooperative game is presented to deal with the condition that selfish secondary users exist.Game theory is used to depict the dynamic cooperative sensing and selfish behaviors of secondary users,and by solving the Nash equilibrium,it allows secondary users to gain the maximum benefit.Through the quantitative analysis of the relationship among spectrum sensing time,energy consumption,collaboration overhead as well as throughput in competitive environment,this paper establishes a non-cooperative game framework among multi secondary users.On this basis,the paper formulates rational utility functions for secondary users,and proves the Nash equilibrium existence for the non-cooperative spectrum sensing game.In order to improve the convergence speed for parameters,a distributed spectrum sensing algorithm is proposed.Experimental results show this method can inhibit selfish behaviors of secondary users,improve the system throughput and the performance of spectrum sensing effectively.Secondly,a distributed cooperative spectrum sensing approach based on trust game is proposed to overcome adverse effects for cooperative spectrum sensing because of malicious behaviors of certain secondary users.The basic idea is to use reputation status parameters to quantify and describe the behavioral characteristics of malicious secondary users,then,adjust reputation status according to strategy selection of secondary users in each interaction.The approach encourages secondary users to choose positive and honest behavior strategies for greater and long term benefits.To highlight the importance of continuing to provide honest services,a flexible reputation mechanism is presented.It puts the continuity of user behavior as an important factor for reputation evaluation,and establishes a differentiation punishment mechanism to respond malicious behaviors of secondary users.The mechanism uses different punishments for "first offender" secondary users and the "recidivist" secondary users,incentives malicious secondary users repented,improves the transaction success ratio among secondary users participated in spectrum sensing,and then guarantees the fairness of cognitive wireless networks and spectrum access performance.Thirdly,in order to improve QoS of secondary users and promote the real performance of cognitive radio networks,spectrum migration scheme based on fuzzy logic is proposed.Without interfering with the primary users,this section aims at the longest occupation time to single spectrum hole,the least spectrum migration times and the shortest spectrum migration decision time for secondary users.In addition,this section defines spectrum migration factors as spectrum characteristic metrics for spectrum migration decision,and establishes a spectrum migration frame based on fuzzy inference,uses pre-decision aid to reduce system complexity and improved spectrum migration efficiency.To shorten spectrum migration decision time and search for suitable spectrum holes,interval Mamdani fuzzy inference based on Mamdani fuzzy inference is proposed,and the twice judgments are used to reduce the complexity of the algorithm.Experimental results show the proposed approach can inhibit the upward trend of forced termination probability,service retransmission probability and average migration times effectively,and improve the effective utilization of CRNs spectrum resource significantly.Finally,in order to improve the performance of access network and terminal services,an intelligent access selection method based on fuzzy neural network is presented.By establishing fuzzy neural network that meets specific needs,network bandwidth,network delay,load balancing at network side and moving speed at terminal-side are used as access factors determined the pros and cons for access network.Then,fuzzification of access factors above by fuzzy logic is done,and fuzzy neural network is employed to inference.At last,the optimal access network can be selected.To improve the performance of parameters learning,a fuzzy particle swarm optimization algorithm is proposed.Compared to the basic particle swarm optimization algorithm,fuzzy particle swarm optimization can overcome the disadvantages that convergence speed slows down even tends to stagnate in the latter,ensure parameter optimization quality for fuzzy neural network.Experimental results show the intelligent access selection method can reduce terminal access blocking rate and packet loss rate,as well as increase the average throughput of access network effectively.