Research On Spectrum Sensing And Access Strategy For Cognitive Radio With Full Duplex Capabilities

With the rapid development of wireless technologies,it is hard to meet a fast growing demand for supporting an increasing number of convenient wireless applications with limited spectrum resource.The contradiction between the scarcity of spectrum resources and increasing demands of wide frequency-bands becomes more and more serious,which restricts the growth of wireless communications.On the other hand,low spectrum utilization is the main cause of spectrum scarcity.Cognitive radio has emerged as a key enabler for spectrum sensing and spectrum access about how unlicensed users find spectrum holes and dynamically share the licensed spectrum efficiently.By allowing secondary users to share the radio spectrum,spectrum sensing and spectrum access technologies have improved the spectral efficiency.By exploring full duplex techniques,a radio system offers the potential to achieve simultaneous sensing and transmission and spectrum utilization can theoretically be doubled compared with the conventional half duplex systems.We explore full duplex techniques in cognitive radio networks to prevent waste of spectrum resource.The main content can be outlined as follows:(1)This thesis analyses spectrum sensing and spectrum access in full duplex cognitive radio networks(CRNs).Most existing CRNs deploy half duplex radios.Existing full duplex spectrum sensing models are explored.Several techniques for spectrum sensing and spectrum access are introduced with their own strength and weakness.It is considered impractical because signal leakage from local output to input,referred to as self-interference.We use Nakagami-m channel to simulate the interference channel since its good fitness.Existing spectrum sensing system model for full duplex is analyzed for further research.(2)We derive closed-form solutions of false-alarm probability and detection probability in Non-Time-Slotted CRNs since primary users may change their state at any time.Sensing performance for full duplex based on energy detection is considered.With full duplex radios equipped at secondary users,the secondary users can operate in either simultaneous transmit-and-sense or simultaneous transmit-and-receive modes.This thesis presents a periodically alternate access strategy(ATAS)for cognitive radios with full duplex capabilities.We analyze the system performance and propose the optimal mode-selection strategy that maximizes secondary throughput subject to a constraint on the given outage probability.Sensing performance is analyzed and the results indicate the validity of the proposed strategy.(3)Matched filter detection outperforms energy detection when prior information of primary signal can be obtained.We consider performance of matched filter detection for full duplex in Non-Time-Slotted CRNs.Closed-form solutions of false-alarm probability and detection probability are derived.Due to the correlation between communication quality and state of secondary users.An adaptive operational mode-selection strategy for spectrum access(AMAS)is proposed.We formulate the problem as a partially observable markov decision process(POMDP)to maximize vacant spectrum usage.Numerical results are provided to verify the proposed protocol and the theoretical results.