Modeling And Analysis Of Performance Bounds For The Transmission Of Secondary User In Cognitive Radio Networks

The rapid development of wireless communication technology prompts the emergence of advanced networks,including Internet of Things,smart grid and Internet of Vehicles,etc.The explosion of variety of wireless equipment brings great pressure to the scarce spectrum resource.Since cognitive radio is capable to permit secondary users to transmit over the spectrum holes of primary users,it becomes one of the key technologies to alleviate the conflict between the lack of spectrum resource and the demand for spectrum.Nowadays,the studies of cognitive radio networks have been focused on the improvement of performance of secondary users on the premise of guarantee the performance of primary user.The existing researches are performed from the perspective of the expectations of secondary users' performance,however,the stochastic nature of cognitive radio networks makes the performance deviate from the expectations.Therefore,literatures investigate the performance bounds of secondary users,yet,there still lacks systematic results.Due to the reasons,this thesis concentrates on the studies of performance bounds for secondary users,and constructs models for the key technologies of the transmission of secondary users.A joint analysis is performed to investigate how the two vital performance parameters,i.e.,throughput bound and delay bound,are influenced by the factors including primary user's state,spectrum sensing errors,multiple access,channel fading and multi-hop transmission,etc.Finally,the optimization method of network parameters is proposed based on the constraint of delay bound.The main contributions of this thesis are summarized as follows:Firstly,performance bounds of multiple access schemes for the transmission of secondary users are investigated.An optimization method of parameters is proposed based on delay bound,which provides a basis for the analysis of performance bounds for the transmission of secondary users.For the contention-based access and contention-free access of secondary users,the thesis analyzes the stochastic characteristics of the dynamic access to the spectrum holes of primary user,and constructs boundary models represented by service curves.Based on the boundary models,the investigation is performed to reveal that how primary user's state and multiple access scheme influence the performance bounds of secondary users.The theory and simulation analysis indicates that with the increase of the number of secondary users,the delay bound of contention-free access scheme increases linearly,while the delay bound of contention-based access scheme increases exponentially.The analysis shows that the delay bound of contention-based access scheme quickly deteriorates along with the increase of the number of secondary users.Accordingly,the thesis proposes an optimization method based on delay bound.Secondly,this thesis studies the performance bounds for multi-hop transmission of secondary users,and proposes a number-of-secondary-users-driven based multi-hop strategy to improve the end-to-end delay,which provides a theoretical basis for the optimization of multi-hop transmission for secondary users.For the existed researches on the boundary models,the correlation between traffic and service is usually considered for the simplification of performance bounds analysis.In this study,an impairment process,independent with traffic,is employed to construct boundary models to obtain a tighter delay bound.As to the limitation of traditional node-by-node analysis for the performance of multi-hop transmission,this study proposes a series equivalence based end-to-end boundary model to get a tighter end-to-end delay bound.According to a further study,it is found that the scaling law between end-to-end delay bound and hop count follows a linear relationship.The theory and simulation analysis indicate that the influence on the performance bounds caused by primary user's state and number of secondary users is stronger than that caused by spectrum sensing errors.Moreover,the end-to-end delay bound has a linear relationship with hop count and an exponential relationship with number of secondary users.Based on the different relationships,the thesis proposes the number-of-secondary-users-driven based multi-hop strategy which improves the end-to-end delay performance.Finally,the investigation on the performance bounds of secondary users' transmission under fading channel is carried out and a fading-sensitive based multi-hop strategy is proposed to improve the end-to-end delay,which provides a theoretical basis for the optimization of the performance of secondary users' transmission under fading channel.Since the Shannon's Theorem is capable to precisely represent the service of fading channel,based on which,a boundary model is constructed from the perspective of transformed domain to describe secondary users' transmission over the spectrum holes of primary user.The thesis analyzes the impact on the end-to-end delay bound caused by channel fading and hop count.The studies indicate that the end-to-end delay bound has a negative exponential relationship with channel fading and a linear relationship with hop count,according to which,this thesis proposes the fading-sensitive based multi-hop strategy which decreases the end-to-end delay.In conclusion,this thesis focuses on the boundary model for the transmission of secondary users,analyzes the performance bounds and proposes the performance bound based optimization methods for network parameters.The studies provide a theoretical basis for the implementation of cognitive radio on the advanced networks.