This paper considers a cognitive radio-assisted wireless information and power transfer system consisting of multipair of transceiver in primary network and 2-hop relaying link in secondary network. In this investigation, a decoded-and-forwardassisted relay node and power splitting protocol are deployed to obtain ability of wireless energy transfer. The relay node harvests energy from the radio frequency signals of the secondary transmitter and primary transmitters in data transmission to the destination by reusing the licensed spectrum resource. We propose 2 policies for wireless power transfer at the relay, namely, (1) multisource power transfer and(2) single-source power transfer. To evaluate performance under energy harvesting regime, we derive the closed-form outage probability expressions and achievable throughput of the secondary network in delay-limited transmission mode. In addition, we investigate the impact of various system parameters including number of primary transceivers, primary outage threshold, and position arrangement of nodes in primary transceivers on the outage performance of the proposed scheme. Furthermore, we evaluate the system energy efficiency to show trade-off metric of energy consumption and throughput. Performance results are presented to validate our theoretical derivation and illustrate the impacts of various system parameters.An important result is that the secondary network is more beneficial than harmful from the primary interference under power constraint and reasonable node location arrangement. , The transmit power constraint of ST by primary outage; P 2 , The transmit power constraint of RL by primary outage; P ST , The maximum allowed transmit power of ST; P RL , The maximum allowed transmit power of RL; MS, Multisource; SS, Single-source.for application development in green communication. [1][2][3][4][5] Energy harvesting has been considered as promising solution to prolong the lifetime in energy constrained wireless network. Moreover, it plays a significant role in cooperative Int J Commun Syst. 2017;30:e3372.wileyonlinelibrary.com/journal/dac