Research On The Optimization Of Efficient Spectrum Sensing And Spectrum Access In Cognitive Radio Systems

With the rapid development of wireless communication technologys,spectrum resources are in short supply.Cognitive Radio(CR)is one of the most promising technologys to solve the spectrum scarcity.For different cognitive radio systems,different performance indexes may be concerned.The energy efficiency is always studied in the systems with energy-constrained nodes.In the systems with high data transmission efficiency,to improve the throughput is usually concentrated.In this thesis,the schemes to improve the energy efficient of CR systems are researched from the perspective of spectrum sensing,and the schemes to improve the system throughput are studies from the point of spectrum access.The main contents and achievements are as follows:1)The effect of the sensing-transmission time slot structure in MAC-layer on the energy efficiency of cognitive radio systems is studied,thus an energy-efficient optimization scheme of spectrum sensing time is proposed.In the scenario of CR system with multiple channels and multiple cognitive users,the energy efficient of the cognitive radio system is modeled,by considering the effect of cooperative spectrum sensing,spectrum handoff and data transmission on the energy efficiency of cognitive radio systems.Then the mechanism of how the energy efficiency being controlled by the sensing time is explored.The optimization problem of energy-efficient sensing time is set up.By optimizing the spectrum sensing time,the energy efficiency of cognitive radio system reaches maximum.The theoretical and simulated results show that there also exists a tradeoff between the spectrum sensing and energy effieicent for the multi-user cooperative sensing system under the switching mechanism.When the system parameters are given,an optimal sensing time,which enables the system to achieve the maximum energy efficiency,can always be found.2)The relationship between spectrum access in MAC-layer and system throughput is studied,thus the optimization algorithm of spectrum access based on partial observable Markov decision process theory is proposed.In the scenario of CR system with single channel,POMDP greedy access algorithm is proposed.It uses the predictive value of the channel state to select behaviors,thus saving some spectrum sensing time and improving the system throughput.In the scenario of CR system with multiple channels,the POMDP suboptimal access algorithm is introduced,and POMDP two-step access algorithm is proposed to improve the POMDP suboptimal access algorithm.The latter improves the accuracy of the spectrum access and the system throughput by stepwise optimization of sensing channels selection and access channels selection.Combined with the spectrum aggregation technology,POMDP adaptive access algorithm is further proposed,which allows the cognitive user to adaptively adjust the number of sensing channels according to the system state and aggregate all the idle channels by spectrum aggregation technology to transmit data,thus improving the system throughput.The simulation results show that the POMDP adaptive access algorithm improves the throughput compared with the existing algorithms.The above work provides some theoretical and technical support for the design of efficient MAC protocol for cognitive radio systems.