| Owing to the technical breakthroughs in high-performance terminal processor and portable operating system, intelligent terminals have been widely used in dif-ferent places and changed our way of life absolutely. Nowadays, services in cater-ing, traveling, and the third-party payment are becoming information-based and networking-based, and their market scale expands continuously which brings more and more profits for the investors. However, the present mobile communication network can not meet the increasing information demands of the users. Thus, the 5th generation mobile communication (5G) has attracted an upsurge of research interests under the impetus of the huge market demand.Nowadays, the key technologies of 5G have been attracting intensive interests but there is still no agreement on which technology should be adopted. Among the candidates, the massive MIMO wins the favor of researchers and operators due to its high spectrum efficiency and energy efficiency and will be one of the key technologies of 5G eventually. Firstly, the large-scale antenna array can concentrate the energy in a much narrower beam, which improves the spectrum efficiency and energy efficiency. Secondly, the interference can be suppressed by a simple linear processing due to the asymptotic orthogonality of the channel vectors. In other words, the extra antennas reduce the complexity of precoding and decoding. Further, the fast fading of channels is averaged out due to the channel hardening of the large-scale antenna array.However, the massive MIMO technology can not guarantee the high performance of 5G alone. Many researchers combine massive MIMO with other candidate tech-nologies such as small cells, Full-duplex and D2D, and then reap all their benefits to satisfy the high performance requirements of 5G. In this thesis, we study the energy efficiency of the massive MIMO system, and optimize the network parameters.First of all, we investigate the energy efficiency for the full-duplex massive MIMO system. Given the transmit powers of both the uplink and the downlink, the closed-form solutions of the optimal number of antennas and the maximum energy efficiency are achieved in the high regime of the signal-to-noise ratio (SNR). It is shown that the; optimal number of antennas and the maximum energy efficiency gets larger with the increase of user numbers. To further improve the energy efficiency, an optimization algorithm with low complexity is proposed to jointly determine the number of antennas and the transmit powers of both the uplink and the downlink. It is shown that, our proposed algorithm can achieve the system performance very close to the exhaustive search.Secondly, both the spectrum efficiency and the energy efficiency are investigated for an uplink massive MIMO system coexisting with an underlay D2D system. The outage probability and the achievable rates of the cellular users and the D2D link are derived in closed-form, respectively. Constrained by the SE of the D2D link and the cellular users, the energy efficiency of the massive MIMO system is maximized by jointly optimizing the transmit power of cellular users and the number of BS antennas. An algorithm with low complexity is proposed to solve the optimiza-tion problem. Performance results are provided to validate our derived closed-from results and verify the efficiency of our proposed scheme.And then, we propose a simultaneous wireless information and power transfer scheme based on the power-splitting technique for the massive MIMO downlink. In the downlink of the massive MIMO, the BS broadcasts the RF signal to the nodes, and each node splits part of the received signal for the information decoding, while the remaining part is used for the energy harvesting. Using the harvested energy, each node can transmit its pilot signal to the BS for the channel estimation. Thus, the network lifetime can be significantly prolonged due to the harvested energy. The ergodic achievable rates of the nodes are derived in closed-form. To maximize the minimum achievable rate among all the nodes, an iterative algorithm with low-complexity is proposed to jointly optimizing the power allocation coefficients for the BS and the power-splitting ratios for the nodes. For the given power allocation coefficient, in each iteration, the optimal power-splitting ratios can be obtained using the closed-form expressions. In addition, the optimality of the proposed algorithm is proved theoretically. Finally, simulation results are presented to validate our closed-from approximations and verify the efficiency of our proposed algorithm.Next, we investigate the energy efficiency of a decode-and-forward relay net-work with wireless energy harvesting. The data transmission between the multi-antenna source node and the destination node is assisted by an intermediate relay node, which operates with the wireless energy harvesting. Through maximizing the instantaneous energy efficiency for the power-splitting based system, the opti-mal power-splitting ratio and the optimal transmit power of the BS are derived in closed-form for each time block. To maximize the instantaneous energy efficiency for the time-switching based system, an optimization algorithm is proposed to ju-diciously determine the harvesting time and the transmit power of the BS for each time block. Furthermore, we optimize the number of antennas of the BS by tak-ing into account the circuit power consumption, as it cannot be ignored when the number of antennas of the BS is large.Finally, we study the multi-pair two-way relay network consisting of two groups of user equipments who want to exchange information through a common relay with massive antennas. In the multiple access control (MAC) phase, the user equipments transmit information to the relay using the harvested energy of the last time block, and the relay decodes the information using the zero-forcing (ZF) or the maximum ratio combining (MRC) technique. In the broadcasting (BC) phase, the relay per-forms broadcasts the RF signal to the user equipments, and each user equipment receives the energy and information using the power-splitting scheme. Due to the channel hardening effect of the large-scale antenna array, the harvested energy of each user equipment in each time block is asymptotically constant.That is, the re-laying networks is a steady system given the power-splitting ratios. We derive the closed-form achievable rates of the user equipments in both the MAC and the BC phases. The power-splitting ratios of the user equipments are optimized to maxi-mize the achievable rates of all the pairs for the ZF or MRC based relaying. For the ZF based relaying, the multi-objective optimization problem (MOOP) is solved in closed-form. For the MRC based relaying, an iterative Pareto improvement algo-rithm is proposed to solve the MOOP.
|
PDF Full Text Request