| Driven by the emergence of new wireless devices and applications, and the increasing interest of wireless services, demand for the radio spectrums has increased dramatically. Cognitive Radio, which has transferred the fixed spectrum assignment policy, is considered as the optimal techniques to realize the dynamic spectrum management. After the convert from analogue to digital transmission, large portions of spectrum in the TV spectrum become available on a geographical basis. Therefore, The IEEE 802.22 Working Group has defined an air interface (PHY and MAC) standard for CR to access to TV white space (TVWS). Interference from CR network, as a main factor to restrict the performance of wireless telecommunication systems, has been concerned throughout the world.This paper mainly contains the following contents:First, the survey of interference problem in CR Network has been given, and two methods to deal with aggregate interference from CR have been proposed.Second, a statistical model of aggregate interference based on in-band spectrum-sensing and opportunistic spectrum access has been developed. In particular, distribution of the aggregate interference is characterized in terms of parameters such as sensitivity, distance between CR network and licensed network, and density of the cognitive radios as well as the underlying propagation environment.Third, based on IEEE 802.22 standards for WRAN and Markov Decision Process Theory, intelligential management has been discussed to deal with aggregate interference from CR network. A form of real-time reinforcement learning, known as Q-Learning, has been proposed to manage the aggregate interference generated by multiple WRAN cells. Meanwhile, intelligential management module has been introduced to treat aggregate interference concentratively. Simulation results reveal that the proposed approach is able to control aggregate interference under constrains, as well as to adjust transmitted power of WRAN cells adaptively. |
PDF Full Text Request