Research On Energy Efficiency Optimization For IMT-Advanced Wireless Networks

Thanks to the adoption of Multiple-Output Multiple-Input (MIMO) combined with Orthogonal Frequency Division Multiplexing (OFDM) based link transmission, Orthogonal Frequency Division Multiple Access (OFDMA) and Carrier Aggregation (CA), IMT-A (International Mobile Telecommunication-Advanced) wireless networks can provide high data rate. However the dramatic increment of energy consumption is followed. The high energy consumption has become the technological bottleneck of IMT-A wireless networks, resulted in the business predicament, impeded the realization of energy saving and the reduction of carbon emissions for mobile communication industry. In order to reduce energy consumption and improve network energy efficiency, improvements and some new designs of key technologies must be provided.According to the different role of key technologies played in IMT-A wireless networks, the research of EE optimization for IMT-A wireless networks includes EE optimization for MIMO and OFDM links, energy-efficient network management, radio resources management (RRM) and network planning for OFDMA networks and networks with CA. Many fruits have been gotten in these fields. However there still several issues should be further studied and some issues have not been involved, which include the EE of Multi-user MIMO (MU-MIMO) link should be further improved, the lack of knowledge about the feature of energy resources is common problem for exsting energy-efficient RRM in OFDMA networks, the EE optimization for networks with CA does not get enough attention.To solve these existential problems, the EE optimization for IMT-A wireless networks is addressed in this dissertation. It is in part supported by the National Science Foundation of China, the fundation from Huawei Company, the National Science and Technology Major Project. The main contents and contributions of this dissertation are issued as follows:1. This dissertation first gives the summary of IMT-A wireless networks and previous researches of EE optimization for wireless networks. Then the challenges and research issues of EE optimization for IMT-A wireless networks are analyzed and summarized, respectively. Lastly, The research situation of EE optimization for IMT-A wireless networks is classified. The shortcomings and issues which are not studied are also proposed on the basis of detailed analysis.2. To improve the EE of MU-MIMO link, the existing researches usually eliminate all inter-user interference by zero-forcing beamforming algorithm, and then focus on energy-efficient power control. When the inter-user interference is all eliminated, the degree of freedom for energy-efficient power control is reduced and the performance of EE is limited. To improve the EE of uplink MU-MIMO links, a convex approximation combined with geometric programming based uplink Energy-Efficient Joint Beamforming and Power Control (EEJBPC) algorithm is proposed. For the EE optimization of downlink MU-MIMO, the EEJBPC problem is more difficult to solve than uplink since the signal to interference plus noise ratio (SINR) of every user is coupled with other users’beamformers. To improve EE of downlink MU-MIMO links, the downlink EEJBPC algorithm is designed on the basis of the duality of MU-MIMO systems. Analyses and simulation results show the proposed algorithms can obtain better EE than previous algortihms and have a good convergence.3. To address the problem of lack of knowledge about the feature of energy resources in previous RRM for OFDMA networks, we first mine the feature of energy resources, i.e. the available energy resources for user is depend on its battery power in uplink networks and energy resources consumed to transmit data for every user are all provided by base station. Then the impact of that feature on energy-efficient RRM is analyzed, and he conclusion that battery power difference among users must be taken into account in fairness of energy-efficient RRM for uplink networks, the fairness of data rate, EE and spectrum efficiency should be considered comprehensively in energy-efficient RRM for downlink networks. Lastly, a battery-aware fairness model is proposed for energy-efficient RRM of uplink OFDMA networks. A battery-aware fairness based energy-efficient power control algorithm is designed to verify the performance of propos-ed model. The simulation results show that the proposed model and algorithm obtain similar network EE as existing algorithms, meanwhile the average lifetime of user’s battery and the users’degree of satisfaction on its lifetime on the basis of its battery power are improved significantly. For downlink OFDMA networks, a model is proposed to realize Energy Efficiency-Spectrum Efficiency (EE-SE) trade-off with fair data rate consideration and an optimal algorithm is designed to solve the problem. Simulation results show that the proposed algorithm can improve EE as well as spectrum efficiency while ensure the fairness of data rate.4. The energy-efficient network planning for networks with CA should overcome some new challenges, which includes limited hardware capabilities, the difference of transmission performance and available bandwidth among carriers. Taking into acount these new challenges, the energy-efficient network planning algorithm for networks with CA is proposed in this dissertation at the first time. We first propose a resource model with considering both maximum equipped number of radio frequency chain and the maximum bandwidth of radio frequency chain. The power consumption model of carrier-aggregated base station is also given. On these basis, we formulate the problem of energy-efficient network planning for networks with CA. Then a random-start two-step Tabu search algorithm is developed to solve the formulated NP-hard problem. The proposed algorithm can search from an infeasible initial solution and reduce the high complexity resulted by increased number of carriers. Through analyses and simulation results, proposed algorithm obtain an impressive improvement of EE with low-complexity and have a good convergence.5. Since the energy-efficient RRM for single-carrier networks does not support carrier configuration and the conventional RRM for carrier-aggregated networks is not energy-efficient, we investigate the energy-efficient RRM for carrier aggregated networks. Since the energy-efficient subcarrier allocation of single-carrier networks can be easily extended to carrier networks with CA, we focus on improving EE by energy-efficient carrier configuration and power control. An optimal energy-efficient carrier configuration and power control algorithm is first proposed. And then a low-complex suboptimal algorithm is designed by exploring the inherent features of carrier configuration and novelly dividing the power control problem. According to simulation, the EE of proposed algorithms are much better than that of conventional algorithm and the suboptimal algorithm can greatly reduce complexity with little loss of EE.