Research On Resource Allocation Algorithms Of D2D Communication Based On Energy Harvesting

With the increase of wireless communication equipment such as smart phones and tablets,and the increasing demand for user traffic,the traditional cellular network is faced with the challenges of shortage of spectrum resources and high energy consumption.D2D(Device-to-Device)communication can reduce the load of Base Station,the transmission power of users and transmission delay.By reusing the spectrum resources of cellular users,D2 D communication can improve spectrum utilization.Energy harvesting technology can harvest energy from the surrounding environment,solve the energy supply problem of low-energy consumption equipment,and realize green communication.Therefore,D2 D communication based on energy harvesting has become a research hots-pot.This thesis mainly studied the resource allocation of energy harvesting D2 D communication in heterogeneous networks.Through reasonable spectrum resource allocation and power control,the interference between D2 D users and cellular users can be coordinated,and the harvested energy can be fully utilized to improve the communication performance of the system.The main research contents and innovations are as follows:(1)A power control and channel allocation algorithm based on non-cooperative game is proposed for the scenario in which the base station supplies power to the D2 D transmitters by means of wireless energy transmission.Assuming that multiple D2 D users can reuse the same channel resource,firstly,the heuristic dynamic clustering method is used to cluster D2 D users and those in the same cluster can share the same channel.Then,D2 D users in the same cluster are modeled as non-cooperative game,the expressions of D2 D users' transmission power and energy harvest time are derived by using KKT condition,and the optimal transmission power and energy harvest time are allocated to D2 D users by the joint iteration optimization method.Finally,KM algorithm is used to achieve the optimal matching between D2 D clusters and cellular users' channel resources to maximize the total capacity of D2 D users.Simulation results show that the proposed algorithm can effectively improve the system performance.(2)A neural network based channel allocation and power control algorithm is proposed for the scenario where the D2 D transmitters use NOMA technology to communicate with two D2 D receivers.Assuming that the D2 D transmitter harvests energy from the surrounding environment,and the system information of all time slots is known,the optimal offline allocation of channel andpower is realized by using the Lagrangian multiplier method under the constraints of users' Qo S and energy causality.Then,the offline optimization results are taken as the training data set to train the neural network to obtain the optimal model of the transmission power.The online power allocation optimization algorithm is further proposed.Numerical results indicate that the offline algorithm can improve the total capacity of D2 D groups,and the performance of online algorithm is close to the offline algorithm.(3)An offline resource allocation optimization algorithm and an online power allocation optimization algorithm are proposed for the the scenario where energy harvesting D2 D transmitter acts as a relay to assist cellular communication while communicating with D2 D receiver by adopting NOMA technology,the outage probability expression of the user is derived.First,assuming that the full system information is available,the channel allocation and relay selection are completed by KM algorithm,and the optimal offline power allocation of D2 D users is realized by reinforcement learning method.Then,the offline optimization results are taken as the training data set to train the neural network,an online power allocation optimization algorithm is proposed according to the obtained transmission power optimization model.Numerical results verify the accuracy of derived outage probability expression and show that the proposed resource allocation algorithm can effectively improve the performance of the system.