| In recent years, Spectrum sharing and cooperative diversity are the two key technologies for cognitive radio networks, cooperative relay networks and cognitive relay networks, which technologies had a profound impact in academia and industry. The research results show that Radio resource management is one of the important methods to improve capacity and spectral efficiency. So we respectively research the theorems and methods of radio resource allocation in cognitive radio networks, cooperative relay networks and cognitive relay networks, in order to address the following problems for the future mobile communication system: (1) How to improve system capacity and Spectral efficiency for the new generation mobile communication network. (2) How to improve spectral efficiency of cell edge. (3) How to improve resource utilization of the spectrum. (4) How to reduce or avoid signal interference for the new generation mobile communication network.In Chapter 2, we study the three issues, including power allocation, the joint radio resource allocation and dynamic spectrum allocation in cognitive wireless networks. (A) We consider the power allocation strategies in the cognitive radio (CR) system with imperfect channel state. As the user has different channel condition in CR systems. We proposed some new power allocation schemes. The targets are to minimize the maximum power allocated to CR users, to maximize the minimum signal-to-interference-plus-noise ratio (SINR) among all CR users and to minimize the maximum outage probability over all CR users. The first power allocation scheme can be formulated using Geometric Programming (GP). Since GP problem is equivalent to the convex optimization problem, we can obtain the optimal solutions for the first scheme. The latter two power allocation schemes are not GP problems. We propose iterative algorithms to solve it. Simulation results shows that proposed schemes can efficiently guarantee the fairness of CR users under the QoS constraint of the primary user and CR users.(B) A general game model base on VCG mechanism for dynamic spectrum allocation is proposed, in which affecting the performance of the dynamic spectrum allocation problem are considered, such as the self-interested behavior of individual users, dynamic change of spectrum, distributed nature of the multi-user spectrum allocation, the VCG mechanism can effectively improve dynamic spectrum allocation and prevent from the cheat behavioral. In order to further improve the performance, we suppose that secondary users observe historic information, predict the bidding policy, a POMDP reinforcement learning algorithm is developed. The simulation results manifests that the proposed learning algorithm can significantly improve the performance of dynamic spectrum allocationIn Chapter 3, we study resource allocation in cooperative relay networks: (A) in order to satisfy each user'QoS requirements, we propose a novel power allocation model to minimize the sum of gap between its achievable rate and required rate for each user, and propose two power allocation algorithms to obtain the optimization solution, which are respectively Sequential Power Adjustment Algorithm (SPAA) and Gradient-Based Power Allocation Algorithm(GBPAA), in our algorithm, each user can achieve their required rate and satisfy their QoS requirements. Secondly, due to limitation of power resource and each users'QoS requirements, a novel joint power allocation and admission control based on energy efficiency is proposed, then, we propose two admission control algorithms to decide which user cannot admit into the relay network, which algorithms are respectively one-step removal admission control algorithm based-on maximum energy efficiency (OSRAC) and step by step removal admission control algorithm based-on maximum energy efficiency (SSRAC). The numerical results demonstrate the effectiveness of the proposed approaches and reveal that our algorithm is energy-efficient. (B) we consider cross-layer design frameworks and performance comparison of HARQ schemes in cooperative relay networks with imperfect channel state information (CSI). Firstly, we propose a state transition models to analyze and calculate the HARQ average packet error rate (PER) and latency performance over Nakagami-m channel with imperfect CSI for relay networks; Secondly, By combing the adaptive modulation and coding (AMC) at the physical layer with hybrid automatic repeat request (HARQ) at data link layer, a cross-layer optimization model that maximizing the average spectral efficiency is formulated under average transmitting delay constraints, and an iteration algorithm for the cross-layer optimization model is designed. We compared the average packet error rate (PER) and spectral efficiency of three kinds of different HARQ protocols in our cross-layer optimization schemes. The numerical results show that our schemes can achieve maximum spectral efficiency. The cross-layer design with AMC and type- III HARQ provides a larger spectral efficiency and lower PER with cooperative relay than the other two HARQ protocols.In Chapter 4, we study resource allocation in cognitive relay networks: (A) we proposed the energy efficiency as evaluation criterion in cognitive relay network, and proposed the cross-layer optimization scheme of joint HARQ protocol and power allocation, we derived closed-form expressions of the average energy consumption per bit data for this cross-layer optimization scheme, and proposed a minimization energy efficiency model, which was restricted by the interference temperature constraint in the primary receiver, as well as constraints on outage probabilities of each second link. Further, we designed a distributed algorithm to solve the joint optimization problem by the Lagrangian decomposition method, and proved the convergence of the proposed algorithm. Theoretical analysis and simulation results showed that this cross-layer optimization scheme based on energy efficiency was feasible, and provided a theory for cross-layer optimization from the perspective of energy efficiency. (B) In order not to affect the normal communication of primary and guarantee the QoS requirements of secondary users, for cognitive relay network, we proposed two optimal power allocation models: (1) maximizing the transmission rate of secondary users; (2) minimizing the total power consumption. Theory analysis shows that two optimal power allocation models are conflicting between spectrum efficiency and power consumption. Furthermore, an optimal power allocation model for jointly the transmission rate and the total power consumption in cognitive relay network is proposed. By the Lagrangian dual method, the optimization algorithm for this model was designed. The proposed algorithms achieve the trade-off between transmission rate and the total power consumption by varying the weight. Simulations results show that the proposed algorithm can effective regulate the transmission rate and the total power consumption of secondary users.
|
PDF Full Text Request