Antenna Beamforming for Energy Harvesting in Cognitive Radio Networks Huidong Liu ⋆ , Jin Chen ⋆ , Guoru Ding ⋆† , Theodoros A. Tsiftsis ‡ , and Corbett Rowell ‡* ⋆ College of Communications Engineering, PLA University of Science and Technology, China † National Mobile Communications Research Laboratory, Southeast University, Nanjing, China ‡ Department of Electrical and Electronic Engineering, School of Engineering Nazarbayev University, Kazakhstan * Department of Electronic and Computer Engineering, School of Engineering Hong Kong University of Science & Technology, China Email: brunojames@163.com, chenjin99@126.net, dingguoru@gmail.com, theodoros.tsiftsis@nu.edu.kz, corbett.rowell@nu.edu.kz Abstract—In this paper, a cooperative cognitive radio network (CRN) with energy harvesting capabilities of its secondary users is considered. Specifically, cooperative spectrum sensing and multi-antenna beamforming are employed to improve the sensing performance and the energy transfer efficiency, respectively. In our approach, a homogeneous CRN scenario is studied where the optimal sensing probability of each second user (SU) is obtained to maximize the control center (CC) throughput while satisfying the energy causality and primary user (PU) collision constraints. An iterative algorithm is proposed to obtain the optimal charging time. Numerical results depict that in an energy constrained scenario, cooperative spectrum sensing with beamforming per- forms much better than cooperative spectrum sensing without beamforming in terms of increased system throughput. Index Terms—Cooperative spectrum sensing, energy harvest- ing, energy cooperation, cognitive radio networks, energy beam- forming I. I NTRODUCTION Green mobile networks have received substantial attention from both academia and industry as a promising approach to increasing energy efficiency in response to the growing concerns about operational expenditures. Green networks ad- dress the global environmental cost of using fossil energy to power the cellular infrastructure [1]. Representative works in- clude energy harvesting hardware and devices, energy-efficient communication techniques, energy-aware network architec- ture/protocol design, energy-friendly software applications, and renewable energy sources. Interest with regard to the powering mobile networks with renewable energy sources has increased substantially. Since radio frequency (RF) signals that can carry energy and be used as a vehicle for transporting information at the same time, simultaneous wireless information and power transfer (SWIPT) proposed in [2] becomes an interesting new area of research. The authors in [3] studied the performance limits for multiple-input-multiple-output (MIMO) SWIPT systems and characterized achievable rate-energy tradeoffs for various practical receiver designs. In order to maximize the harvested energy, it is necessary to coordinate the transmit direction to the receiver (energy beamforming). The key for successful energy beamforming is the channel state information (CSI) knowledge at the transmitter side [4]. In addition to the CSI determination, another key design objective in wireless communication is the spectral effi- ciency which addresses the increasing spectrum demands of multimedia services. Cognitive radio networks (CRNs) are seen as promising technologies to achieve greater spectral efficiency [5]. Large portions of the licensed spectrum are seriously underutilized, leading the concept of opportunistic spectrum access, allowing secondary users to exploit under- utilized spectrum gains [6]. Spectrum sensing is a well-recognized enabling technique for CRNs. Due to the hidden terminal problem, a secondary user (SU) may not notice the existence of the primary user (PU) and increase the interference to the licensed systems. One method to solve the above hidden terminal problem is allowing multiple cognitive users to cooperatively perform spectrum sensing. It has been shown that the performance of spectrum sensing can be improved with an increase of the number of cooperative partners [7] and can therefore overcome both the hidden terminal problem and poor channel conditions (multi- path fading and shadowing) [8]. Recently, there has been research on CR systems using energy-harvesting techniques [9]. Optimal couple of sensing duration and detection threshold was studied in [10]. In our previous work, an RF-energy harvesting CRN employed coop- erative spectrum sensing and an optimal cooperative spectrum sensing strategy was proposed to maximize the control center (CC) spectral efficiency [11]. Due to the limited antennas at the power source, however, the energy efficiency based on the aforementioned systems was difficult to satisfy the practical requirement without energy beamforming, especially when for long transfer distances. In this paper, we consider beamforming RF-energy har- vesting CR systems which use cooperative spectrum sensing in order to improve the sensing accuracy and the energy efficiency. The main contributions of this paper are twofold: i) A new mutually beneficial relationship between the CC and the SU is proposed and ii) An optimization problem to achieve the optimal sensing-throughout trade-off under the