Cooperative and Cognitive Communication in Wireless Networks

In the last decade, significant efforts and progress have been made by both the industry and academia to meet the rapidly growing demand for wireless applications and services. To achieve more flexible, dynamic and intelligent use of the limited wireless spectrum, cooperative transmission and cognitive networking are proposed as two of the key technologies for the next generation wireless communication systems, such as Long-Term Evolution Advanced. Cooperative transmission techniques, such as cooperative relaying and Multiple-Input and Multiple-Output (MIMO) can increase spectrum efficiency by utilizing the diversity of wireless channels, while cognitive transmitters tune their transmission parameters according to the environment to optimize network level performance. In this thesis, we provide performance modeling and analysis of different cooperative and cognitive communication techniques to exploit their potential. In the first part of the thesis, we investigate the performance of hop-by-hop cooperative communication on a multihop transmission path applying spatial reuse time division multiplexing, where interference from simultaneous transmissions exists. Based on the models, we compare the performance of hop-by-hop cooperation with the performance of traditional simple multihopping schemes, and give the regimes where hop-by-hop cooperation achieves significant gain. Considering random networks, we propose cooperative geographic routing, the integration of hop-by-hop cooperation with traditional geographic routing, and evaluate the effects of the topology knowledge range and the network density. In the second part of the thesis, we discuss how cooperative transmission techniques can be utilized in cognitive and hierarchical spectrum sharing networks, where the primary users have transmission guarantees, and the coexisting secondary users need to be cognitive and adjust their transmissions in the shared spectrum bands to conform constraints from the primary users. We consider large-scale coexisting primary and secondary networks, where concurrent primary and secondary transmissions are allowed, and the secondary users provide cooperative relaying for the primary ones and control the interference at the primary receivers by tuning the probability of transmitting and by forming a primary exclusive region around each primary receiver within which all secondary users have to be silent. We define a unified analytic framework to model the performance of cooperative spectrum sharing and cognitive transmission control, characterize their achievable gains, and show that both of the networks have strong incentives to participate in the collaboration. Finally, we investigate spectrum sharing networks where both primary and secondary users have stochastic packet arrival. Under the constraint that the performance of primary users does not degrade, we find the dilemma for the secondary users. That is, if a secondary user chooses to cooperate, it can transmit immediately even if the primary queue is not empty, but has additional costs for relaying primary packets, such as increased power consumption. We propose a dynamic cooperation scheme for the secondary user so that it can make sequential decision on whether to cooperate or not in each time slot based on the state of the network. We show that optimal sequential decision is necessary to efficiently trade off the cooperation cost and the packet delay of the secondary user.