Investigation On Link Initialization Control And Optimal Resource Allocation In Wireless Cognitive Radio Networks

With the rapid development of wireless communication technology and the continuous emergence of various applications and services, the conflict between huge demand and the under-utilization of fixed spectrum allocation policy is becoming more and more serious. Therefore, cognitive radio is recommended as a promising technology to solve the above problem, by al-lowing the secondary users to sense the environment and opportunistically access the licensed spectrum without interrupting the transmissions of the primary users. In addition, cognitive radio is a key technology for the next generation heterogeneous wireless networks, it can provide in-telligence to both the mobile user and provider equipments to manage the network efficiently. In this dissertation, link initialization control and optimal resource allocation in wireless cognitive radio networks are intensively investigated, and the contents of this work are listed as follows:Without the common control channel, how to fast and efficiently establish an initial link between the transceiver pair is important and challenging in cognitive radio networks. Two different mechanisms are studied based on single channel hopping and group channel hopping, respectively. For the mechanism based on single channel hopping, the successful probability of initial link establishment and the stability of established link is modeled. Then an adaptive channel hopping strategy based on the primary user's activity of each channel is proposed and showed to have better performance than the mechanism based on random channel selection. Besides, two heuristic approaches are presented. For the mechanism based on group channel hopping, the closed-form expression of the probability of successful link establishment and that of the average number of trials needed to achieve certain success probability are derived. Then two optimization problems aiming to minimize the number of channels to be chosen and the average number of trials under different constraints are formulated. The asymptotic results for conditions under which the link can be successfully established with probability arbitrarily close to 1 are presented.The distributed spectrum access problem with the secondary users'QoS requirements is investigated in cognitive radio networks. The distributed spectrum access in cognitive cellular networks is analyzed with the objective to maximize the number of admitted secondary users under the constraint of interference, while satisfying the secondary users'QoS requirements. Considering the above requirements, it is not always possible that all secondary users can be admitted concurrently, instead, one or more of them have to be removed so as to maximize the number of admitted SUs. Thus, the strategy of gradual removal is introduced into cognitive cel-lular networks and propose two distributed power control algorithms. Then, an initial spectrum access control for cognitive radio networks is studied, which aims to maximize the number of ad-mitted secondary users under the constraints of interference temperature, while providing Active secondary users'QoS Protection (AQP). Here, AQP means that the SINR of all active secondary users will not fluctuate below some predefined thresholds during new secondary users'initial spectrum access process. Therefore, a new distributed power control algorithm is proposed in conjunction with a simple alarm mechanism which ensures the primary users'communication can not be interrupted. Then two realizations of the spectrum access algorithms are presented according to whether the MPs can act as local controllers or not.The cross-layer resource allocation problem in cognitive wireless MESH network based on both the underlay and overlay spectrum access model are studied, respectively. For the overlay spectrum access model, the objective is to minimize the system activation time to satisfy the given traffic demands, under the constraint of multiple access interference and the limited avail-able spectrum bands at each user, by jointly optimizing spectrum allocation, routing and time scheduling. Then a resource allocation algorithm based on column generation approach is de-veloped. For the underlay spectrum access model, the objective is to maximize the total utilities in cognitive wireless MESH networks, by jointly allocating each link's rate, power and subchan-nels under the constraints of interference temperature and multiple access interference. Then, a centralized resource allocation algorithm is developed based on the column generation approach and shown to be optimal. So it can perform as a criterion for designing other algorithms. Be-sides, considering the applicability of algorithm in distributed system, a near-optimal distributed algorithm is proposed, which allocates subchannel based on routing information at first, and then jointly allocates the resource of rate and power.