Energy Efficiency Optimization Of D2D Underlaying Massive MIMO Cellular Networks

D2D(Device-to-Device)communications underlaying cellular networks has been recognized as one of the key technologies for 5G communication systems.D2D users(DUEs)directly communicate with each other by reusing the cellular users'(CUEs)resources,which can greatly improve network spectral efficiency(SE)and energy efficiency(EE).Since the battery life of the mobile DUEs is limited and the resource reuse will cause additional co-channel interference,in order to reduce the interference and improve the system EE,proper resource allocation and power control should be paid more attention to when introducing D2D communications into cellular networks.Another promising technology being recognized as one of the candidates for 5G systems is massive multiple-input multiple-output(MIMO),introducing D2D into Massive MIMO(multiple-input multiple-output)systems has attracted intensive attention since combing the two can further increase the network SE and system energy efficiency(EE).However,introducing D2D into Massive MIMO systems will cause severer channel estimated interference.Therefore,it is important to propose appropriate pilot reuse scheme.In this paper,we firstly study the energy efficiency optimization for D2D communications underlaying traditional single-antenna cellular networks,and then investigate the energy-efficient power control,pilot allocation and pilot length optimization for D2D communications underlaying Massive MIMO systems.The main contributions and innovations in this paper can be summarized as follows;1)The distributed energy-efficient resource allocation and power control(PC)for D2D communications underlaying single-antenna cellular networks.In this paper,a grouping scheme for resource allocation is proposed according to the interference limitations and EE constraints.Based on the grouping scheme,we study how to maximize the EE of the whole system through power control method for each group while guaranteeing the quality-of-service(QoS)of both CUEs and DUEs.The EE maximization problem is formulated as a global optimization problem,which is non-convex and thus is difficult to solve.To solve it,a distributed iterative optimization algorithm is proposed and analyzed.Simulations are done to show the performance of the proposed algorithm.Results show that the proposed algorithm converges very fast and the system energy efficiency is improved significantly.2)The pilot reuse and power control for energy efficiency(EE)optimization on the uplink of D2D communications underlaying massive MIMO systems is investigated.Tofacilitate analysis,the uplink coherent time is divided into two phases:the channel estimation phase and the data transmission phase.During the channel estimation phase,DUEs are allowed to reuse the orthogonal pilots of CUEs,so the pilot overhead extremely decreases compared with orthogonal training scheme.Aiming to reduce the channel estimation error and preliminarily improve the system EE,we design a low-complexity pilot reuse(PR)scheme to select the proper D2D reusers for each pilot.Then,the channel of both user-BS links and D2D links are estimated through MMSE method.In order to reduce the DUE-to-CUE interference,the pilot power of CUEs andthat of DUEs are considered to be different in this paper,and an optimal power ratio is found through intensive simulations.During the data transmission phase,the MMSE detector is equipped at the BS to effectively suppress the DUE-to-CUE interference.Then,motivated by improving the system energy efficiency(EE),a distributed power control algorithm based on non-cooperative game is investigated.Under the quality-of-service(QoS)constraints of both CUEs and DUEs,a Nash equilibrium can be found after sufficient iterations,and the system EE is notably enhanced.3)The pilot length optimization for spectral and energy efficient D2D communications underlaying Massive MIMO networks.In order to reduce the channel estimation error caused by the pilot reuse among CUEs and DUEs,a heuristic pilot optimization and pilot allocation(PA)scheme is formulated through minimizing the channel estimation-interference-strength(CEIS).By taking into account the interference among both CUEs and DUEs,the optimal pilot length is obtained with the purpose of maximizing the system EE and SE.Intensive simulations are done to investigate the relation between the pilot length and the number of D2D pairs in terms of EE and SE.The results show that the optimal pilot length always exists with the variation of the number of D2D pairs and the PA with optimal pilot length can significantly improve the performance of both EE and SE compared with the completely reused or orthogonal pilot scheduling strategy.