Low-Complexity MIMO Systems Design Based On Energy Transmission

ABSTRACT:Simultaneous wireless information and power transfer is a new wireless communication technology, which can obtain energy from the grid or the natural environment, and then transmit energy to the user terminals. So it is used in the short distance transmission scenario in the room, e.g. sensor networks, etc and then and then establish a self sustainable green wireless communication systems. This innovative technology has a broad application prospects. However, there are many problems need to be solved. First, in wireless communication systems, the wireless radiation signal is only used for information transmission, but not energy transmission, so that the source of energy in wireless communication system and the energy efficiency problems need to be studied; second, multi-antenna technology has not been widely used in the system; third, robust design is needed in the system.So we consider the joint optimization of time allocation and precoding design in a wireless powered time-division-multiple-access (WP-TDMA) system, which consists of a power station (PS), several users and a sink node. Assume that each node is equipped with multiple antennas, and the PS has reliable power supply while the users are passive and have to harvest energy from PS by wireless power transfer (WPT). The harvest-then-transmit protocol is adopted where all users harvest energy in the first downlink (DL) timeslot, and then send their own information to the sink by TDMA protocol in the uplink (UL).1) We first formulate the optimal design of sum throughput maximization under a power constraint as a convex semidefinite programming. Then an iterative waterfilling solution with projected subgradient update is proposed to serve as a low-complexity implementation. We prove the optimality of the energy beamforming, which is validated by the special case where each user sends only one data stream to the sink. This further motivates us to propose a suboptimal solution, which can greatly reduce the implementation complexity but incur little performance loss. We also consider the impact of the uncertain channel state information in the WPT phase and solve the problem by iS-procedure method.2) We maximize the minimum throughput and solved it by convex technology. Decentralized algorithm is porposed as following. First, fixing the DL time allocation, power precoding matric, we can obtain the information precoding matrices by water-filling (W.F.) method, DL time allocation and maximum min-throughput by fast fixed-point iteration algorithm. Then, by projected subgradient method the power precoding matric can be obtained. At last the DL time allocation is obtained by golden section search.3) The good performances of these two algorithms have been proved and the results show that many factors may affect the performance of the system.