Key Performance Analysis In Cognitive Radio Networks

With the rapid development of mobile internet, all kinds of new wireless applications are emerging, the fixed amount of spectrum resources are not satisfying the bandwidth requirement of the new traffic. However, a report from FCC shows that more than90%of the allocated spectra remain idle most of the time, which means most spectrum resources are underutilized under the current spectrum allocation policy. Cognitive radio networks (CRNs) are born under such a background. CRNs are based on cognitive radio technology, which can observe the environment, and obtain cognitive information for radio resource management (RRM) and decision. With the interaction of cognitive information, authorized spectra can be allocated flexibly and efficiently, to ease the contradiction between spectrum shortage and its low utilization. CRNs contain primary networks and secondary networks, which have different priorities to access the spectrum, and makes it different from traditional networks on networks architecture, fundamental limits and RRM. Therefore, it is important to analyze the key performance of such networks, which is significant to the optimization and RRM of CRNs. This dissertation analyzes the key performance of CRNs, which includes the following three parts.The first part studies the link capacity of CRNs. From the link perspective, this dissertation analyzes the achievable rate region of the multiple user cognitive multiple access channel (MUCMAC). The MUCMAC is a multiple access channel which includes one primary user (PU) and multiple secondary users (SU), the SUs can obtain the PU’s message through cognition, and access the channel under the overlay paradigm. After obtaining the PU’s message, the SUs can perform dirty paper coding (DPC) and cooperative coding to achieve the rate region. On one hand, SUs can allocate part of their transmit power to aid the transmission of the PU’s message. On the other hand, SUs can use DPC to avoid the interference from the PU. Under such a strategy, the capacity region of the traditional multiple access channels can be enlarged. Using such a coding scheme, we derive closed form expressions of the MUCMAC under the additive white Gaussian noise channel and the Rayleigh fading channel, and analyze the asymptotic performance of the MUCMAC.The second part analyzes the connectivity of CRNs. From the network perspective, we analyze the connectivity of a one-dimensional (1-D) secondary network (SN) and a two-dimensional (2-D) SN. For the1-D SN, we study an overlaid network which includes a2-D PN and a1-D SN. By analyzing the conditions for the connectivity of the SN, we decompose it to two events, and derive the corresponding probability and conditional probability to obtain the closed form expression of the SN’s connectivity. For the2-D SN, we analyze the SN’s connectivity region under the interference limited scenario. Different from most existing researches that adopt Poisson Boolean model to analyze the connectivity, we consider interference as the key performance limiting factor, and study the connectivity of the SN using percolation theory. By mapping and deriving, we prove that the connectivity region of the interference limited SN is the same as that of the Poisson Boolean modeled SN, and concludes that interference will not influence the connectivity region of the SN.The third part studies the network capacity of CRNs. Also from the network perspective, we study the capacities of the cognitive ad hoc network (CAHN) and the cognitive cellular network (CCN). In the CAHN, we first analyze the transport capacity of the SN. Different from current researches that mainly focus on the scaling law, we use the theoretic framework of stochastic geometry to model the communication of the SN, and derive a closed form expression of the SN’s transport capacity. Then we analyze the transmission capacity (TC) of the SN. We induce preservation regions between the PN and the SN, derive a closed form expression of the SN’s TC, and demonstrate that preservation regions can increase the TC of the SN. In the CCN, we analyze the capacity of the SN for both uplink and downlink. For the uplink, we derive expressions for the average link rate and throughput of the SN, and make comparative analysis between regular and random base station deployments. For the downlink, we study the SN’s sum rate of the multiple description coding multicast (MDCM). Based on stochastic geometry, we derive a closed form expression of the SN’s sum rate under MDCM, and demonstrate its superiority compared with conventional multicast.