Research On Resource Allocation In D2D And Energy Harvesting Based Cognitive Relay Networks

With the flexible sensing capability of transmission resources,cognitive relay network can lead to higher resource utilization,and it has been widely exploited in practical scenarios.In Device-to-Device(D2D)communication systems,users can communicate with each other directly without traversing through the base station,which can decrease users' interference range and improve the spectrum efficiency.In energy harvesting(EH)based communication systems,users can obtain transmission energy from the surrounding environment,which can increase the energy efficiency.Thus,cognitive relay networks with D2 D communication and EH technology can further improve resource utilization and network performance,which has been considered a hot topic in recent years.After combining D2 D and EH techniques with cognitive relay networks,an important issue is to efficiently allocate communication resources.Therefore,in this thesis,the features of D2 D and EH based cognitive relay networks are analyzed firstly.Then,according to the sensing capability of different resources and network environments,the resource allocation problems in D2 D and EH based cognitive relay networks are investigated based on optimization theory,game theory and probability network.Several algorithms are proposed in order to improve the resource utilization of the network.The main works and outcomes can be summarized as the following aspects:First,based on the spectrum sharing mechanism,this thesis investigates the resource allocation problem in D2D-enabled cognitive relay networks,where the cellular users(CUs)share the spectrum resource with D2 D users(DUs)in the underlay mode.Considering the network throughput and fairness among different users,several algorithms are proposed to mitigate the interference between CUs and DUs and improve the resource utilization.1)To improve network throughput,idle users are used as relay nodes for DUs.With the minimum rate requirement of all users,two algorithms are proposed to allocate the relays and spectrum resource based on coalition formation game theory.Specifically,the first algorithm is designed based on merge and split rules while the second one is developed based on the single user's movement.Both the algorithms are proved to be stable and convergent.Simulation results show the effectiveness in allocating resources of the proposed algorithms.2)Considering the selfishness of DUs and the minimum rate requirement of CUs,a coalition formation game based algorithm is proposed to deal with the joint optimization of transmission power,time slots and spectrum resource for DUs.Simulation results show that the proposed resource allocation approach achieves a good tradeoff between users fairness and the system sum rates.Then,based on the sensing capability of link state,this thesis proposes a link state prediction based algorithm in D2D-enabled cognitive relay networks with overlay mode.With the probability network theory,the joint mode selection,power control and channel allocation algorithm is proposed to improve the resource utilization while avoiding the transmission link break and co-channel interenrece.Simulation results show that the proposed algorithm can effectively improve the link stability.Finally,based on the sensing capability of energy resource,this thesis investigates the resource allocation problem in EH-enabled cognitive relay networks.Since the randomness of the harvested power may influence the system energy efficiency,this thesis studies the resource allocation problem in the following two scenarios: 1)Considering the randomness of the harvested energy from the nature environment,an queue theory based power allocation scheme is proposed to maximize the throughput of the secondary network in cooperative beamforming based system;2)Considering the harvested radio frequency(RF)energy,a cooperative transmission and RF energy harvesting protocol for multiple data sources is proposed to allocate the time resource for the energy harvesting process and data transmission.Simulation results show the effectiveness of these two schemes and the improvement of energy efficiency.