Research On Resource Allocation Algorithm Of Relay-aided D2D Communication Based On Energy Efficiency Optimization

With the beginning of 5G commercial,the number of mobile devices increases rapidly.The performance of wireless network system would face a crucial test as the data rate requirement becomes harsh in terms of multiple quality of service and needs including capacity,delay and energy efficiency.As one of the promising technologies in the next generation of wireless communication system,device-to-device(D2D)communication which allows two proximity devices communicate directly with reusing the spectrum resource of cellular users,can improve spectral efficiency,reduce end-to-end latency and offload network traffic.The benefits of D2 D communication may be limited due to long distance and inferior channel conditions.In such cases,it is necessary to employ relays so that we can maintain the completion of D2 D communication and enlarge the coverage of cellular networks.However,because of the extra power consumption with introducing relay,how to improve energy efficiency(EE)is crucial for future wireless communication.Therefore,this paper studies the resource allocation for relay-aided D2 D communication underlaying cellular network and improves the EE of total D2 D users.The main contributions are summarized as follows:1.The bottleneck effect-oriented power allocation and relay selection scheme in D2 D two-hop communication: Considering that the capacity of D2 D communication could be limited by the link with lower capacity on account of bottleneck effect,an optimal problem of energy efficiency under the constraints of quality of service(QoS)about D2 D users and cellular users is formulated as a mixed integer nonlinear programming(MINLP)problem,with the goal of maximizing the total EE of D2 D users.Then,the problem is divided into two subproblems,power allocation and relay selection,with being tackled respectively.The power allocation problem is addressed by Dinkelbach method and Lagrange multiplier method with the energy efficiency of the first-hop and the second-hop D2 D links being maximized.By designing an alternative region of a few relays for each D2 D pair to reduce the computation complexity,a novel relay scheme is proposed based on Q-learning.Simulation results indicate that the proposed power control scheme can nearly reach the theoretical maximum of energy efficiency.And the proposed joint power control and relay selection scheme outperforms the existing schemes in term of EE of total D2 D users.2.Joint optimization and resource allocation algorithm with low-complexity.Considering the “three-dimensional” optimization and resource allocation problem which includes power control,relay selection and subcarrier assignment in relay-assisted D2 D communication system,this paper formulates one-to-one spectrumallocation reusing scenarios.Based on that,with the joint research and optimization,on the one hand,this paper can analyze the restrictive relation between each communication part.On the other hand,this paper can realize the optimization in system level.Firstly,assuming the signal to interference plus noise ratio(SINR)of the first-hop and the second-hop D2 D links are same,a power control algorithm is proposed with low-complexity.Then,by designing an alternative region for each D2 D pair and reducing the state space size of Q learning,a relay selection scheme is proposed with low-complexity too.At last,according to the results of power control and relay selection,the matrix of EE between D2 D users and cellular users is calculated and subcarrier assignment problem is tackled by Hungarian algorithm.Simulation results show that the joint power control,relay selection and subcarrier assignment scheme can effectively improve the EE of total D2 D users.