Pilot Allocation And Power Optimization Of Massive MIMO Cellular Networks With Underlaid D2D Communications

Massive multiple-input multiple-output(MIMO)and device-to-device(D2D)communication technologies have played a huge advantage in achieving high spectrum efficiency(SE),which have become the most promising key technologies in 5G and B5 G wireless communication networks and are expected to become important components of 6G.The acquisition of accurate channel state information(CSI)is the basis for achieving high SE,and CSI is obtained by the base station using channel estimation technology.Limited by the coherence time,the pilot cannot satisfy all users of the massive MIMO system to achieve pilot orthogonality.Therefore,users inevitably share the same pilot,leading to a pilot pollution problem that limits the improvement of system performance.Even if massive antennas are deployed in base stations,pilot pollution still limits the system capacity cannot be increased.Therefore,Pilot allocation becomes the research focus of massive MIMO systems.Moreover,a smaller transmission power will cause the signal transmission failure,and a larger transmission power will reduce energy efficiency.Facts have proved that power is another important challenge that affects system performance.In addition,enhancing the cooperation between massive MIMO and D2 D communication can greatly improve the SE,but it will aggravate pilot pollution.In response to the above-mentioned challenges,we first study the pilot allocation and pilot length optimization schemes for multi-cell massive MIMO systems.And then extend the multi-cell massive MIMO system to the D2 D communication underlay massive MIMO(denoted ‘massive MIMO-D2D' for short)cellular system,focusing on pilot allocation and power optimization.The main contributions of this thesis are as follows:1)The pilot allocation algorithm for reducing pilot pollution and improving the SE in the multi-cell massive MIMO system is studied.Considering pilot pollution and coherence time limitation,a joint pilot allocation and pilot length optimization scheme is proposed.First,to avoid users with severe mutual interference from sharing the same pilot,the similarity based on channel coefficients is defined to distinguish pilot pollution between users,and an angle of arrival(AOA)positioning method is designed to calculate the distance used to measure potential power interference between users.Then we assign orthogonal pilots to the users with greater similarity and shorter distances.Secondly,how to plan the pilot length to promote the improvement of system performance is studied,and a lowcomplexity extrapolation and interpolation algorithm is proposed to optimize the pilot length.Finally,a joint pilot allocation and pilot length optimization algorithm is designed to maximize the system SE.Simulation results verify that combining pilot allocation and pilot length optimization scheme can significantly improve the system SE compared with random pilot allocation,smart pilot allocation,and pilot-free length optimization scheme.2)The pilot allocation and power optimization algorithms for improving SE in massive MIMO-D2 D cellular systems are studied.Cellular users and D2 D users share the same pilot aggravates the pilot pollution and power interference of the system.In response to these problems,firstly,we analyze the disadvantages that the existing pilot allocation problem cannot provide fairness and improve system performance simultaneously,and propose a new pilot allocation problem,i.e.,the SE product is maximized for enhancing the system SE and ensuring fairness among user simultaneously.Secondly,the maximum SE product is an optimal combination problem,we propose a competitive pilot gray wolf prey(PGWO)algorithm that can find the optimal SE accurately through a global search to solve this problem.Finally,aiming at the problem of power control and further improving the SE,a maximum minimum fairness problem is formulated,and a bisection convex optimization algorithm is provided to solve this problem.Simulation results show that the proposed SE product problem,the proposed PGWO algorithm and bisection convex optimization algorithm promote fairness for users while further enhancing SE compared to traditional pilot allocation scheme.