Resource Allocation In Future Broadband Wireless Communcations

With the increasing requirements of service and rate, how to make use of the limited wireless resource effectively to enlarge channel capacity and improve system quality has become a critical question to solve for the future wireless broadband communications. In this paper, we focus on two promising wireless communication technologies, i.e. cooperative transmission and cognitive radio, aiming at improving system capacity, increasing the network resource utilization, guarantee QoS requirement of different users and etc. to study on the resource allocation of them. Meanwhile, The optimization theory, game theory, and other marketing methods are employed as academic rudiment to make wide research deeply and intensively.Supported by several project of National Natural Science Foundation of China, National High-Tech Research and Development Plan of China, National International Science and Technology Cooperation Project under Granted of Sweden and Canada, we aim at the resource allocation research in the future broadband wireless communications in the dissertation, by deep analysis of cooperative transmission and cognitive radio technologies, exploiting new solving methods to gurateen the QoS of users and provide high frequency efficiency through optimizing the resource from the aspect of system-level and user level. The main contributions of the dissertation are as follows:(1) We investigate a multi-user multi-service scheduling scheme in the OFDMA based two-hop cooperative relay network, which aims at the cooperative resource allocation in this paper. We propose an adaptive Aggregated Utility Function (AU-Function) as the optimization objective, which simultaneously takes various multi-service QoS requirement and fairness among different services into consideration. We format the utility function into a several QoS parameters captured form, including rate, delay, jitter and Packet Loss Ratio (PLR) of services, which combines the QoS requirement well. Then, the complex resource allocation is decoupled into a joint relay-subcarrier selection and power allocation problem. Simulation results confirm that the proposed algorithm achieves an efficient balance among rate, delay, PLR, etc., and show that the users'QoS can be evaluated adaptively by the full dimensional utility consideration.(2) Cooperation allows wireless network users to benefit from various gains such as an increase in the achieved rate or an improvement in the bit error rate. In the paper, we propose a distributed Hierarchical Game (HG) theoretic framework over multi-user cooperative communication networks to stimulate cooperation and improve the network performance. First, we study a two-user decision making game in the OFDMA based subscriber cooperative relaying network, in which subscribers transmit their own data in the first phase, while helping to retransmit their partner's or choosing to freeride in the second phase. Instead of consulting to a global optimal solution, we decouple the cooperation resource allocation into two level subproblems:a user level Nash game for distributed cooperation decision and a Base Station (BS) level coalition game for centralized resource allocation. In the proposed HG algorithm, where mutual cooperation is preferred and total payoff is transferable, we prove it converges to a unique optimal equilibrium and resolve the subcarrier assignment and power allocation among the couples. Besides, we discuss the existence of the publishing and rewarding coefficients in order to encourage cooperation. Then, we extend the HG to multi-user cases by coupling among subscribers according to the location information. The simulation results show that the proposed scheme with the distributed HG game achieves a well tradeoff between fairness and efficiency by improving the transmission efficiency of adverse users and outperforms those employing centralized schemes.(3) In this paper, we analyze two classical performance metrics for user cooperation with network coding scheme. Two closed formulation of capacity and outage probability for OFDMA based two users cooperation are presented. Compared with the DT and conventional CT, simulation results show the advantages of CT_NC. It provides 11.8% higher link capacity than CT cases averagely; meanwhile, it obtains 50% lower outage probability. Therefore, CT_NC is much more powerful in both the system throughput and robustness due to the network coding diversity than the direct transmission scheme and conventional user cooperative protocol.Besides, a network coding based two-subscriber-cooperation scheme in a full-duplex transmission mode of OFDMA system is proposed to further improve the network throughput. Besides, a double level Nash Bargaining Solution (DL_NBS) game is adopted to resolve the inter-user resource bargaining problem. In the inter-user pair, the pairwise capacity based NBS is utilized to allocate subcarrier and distribute power between the cooperative subscribers. The simulation results show that the proposed CoNC achieves system capacity 49.1% higher than the normal cooperative transmission, and 46.4% higher than the direct transmission. Meanwhile, compared with the traditional resource allocation algorithms, the two-level NBS solution achieves a well tradeoff between fairness and efficiency, as well as perfectly suits the distributed subscriber cooperation scenario.(4) Cognitive networks are designed based on the concept of dynamic and intellectual network management, characterizing the feature of self-sensing, self-configuration, self-learning, self-consciousness, and such. In this paper, focusing on the spectrum sharing and competition, we propose a novel behavior modeling methodology basing on a paradigmatic cognitive radio network. We first discuss the preponderance and challenges of cognitive network, and explore the special behavior features basing on a modeling prototype for spectrum competition in a multi-radio cognitive network. Meanwhile, a rounded OODA cognitive behavior is explicitly given, within which the decision process is mathematically illustrated. We present a behavior model from the economical theoretical perspective, and introduce a double auction decision making algorithm which resolves the spectrum sharing between the primary network and secondary networks subtly. Two different utility functions for primary users and secondary users are designed basing on a supply-and-demand relationship between them. Also, we adopt expectation and learning process in the decide module, which takes consideration of the variance of channels, transmission forecasting, afore trading histories and etc. Numerical results with four bidding strategies are presented to reinforce the effectiveness of the two proposed utility evaluation based decision modules under two scenarios. Besides, we prove the proposed behavior model based spectrum access method maintains comparable frequency efficiency with traditional centralized CR access approaches.In order to fully utilize scarce spectrum resources dynamically and intelligently, we model a cognitive radio network with various selfish autonomous secondary users who strategically interact to acquire dynamically available spectrum opportunities. Our main focus is on developing solutions for cognitive users to make full utilization of the spectrum with each other, given the selfish nature of users in the wireless network and complete for the limited and time-varying spectrum opportunities. To analyze the interactions among users given the environment disturbance, various transmission requirements, we propose a double auction framework to tackle the competition among users for spectrum opportunities over time. The double bidding actions affect the resource allocation and, hence, the rewards and future strategies of all users. Based on the observed resource allocations and corresponding rewards, reinforcement Learning methodology is deployed by wireless users to improve their bidding policy, building on which we formulate bidding prices for the secondary users and asking prices for the primary users. Simulation results show that the proposed Q-learning based double auction algorithm can significantly improve users'own bidding strategies and, hence, their performance in terms of packet loss, bidding efficiency and transmission rate is improved progressively.Summarily, the dissertation do hard work on wireless resource allocation problem in cooperative transmission and cognitive radio network to improve system capacity, increase the network resource utilization, guarantee QoS requirement of different users and etc. The OFDM transmission is included in the physical layer, many mathematical methord are employed, and the cross-layer theory are taken as a methodology, this dissertation studies on allocation of the multiple network resource, i.e. channel, power, time, frequency etc. Therefore, it provides a deep and wide research theoretically for the implement and optimization of cooperation transmission and cognitive rado technologies in the future wireless communication.