In cognitive radio (CR) relay networks,most previous work concentrates on maximizing physical layer quality of service (QoS),e.g.,spectral efficiency and achievable data rate,as relay selection criteria.However,the residual energy state of relay nodes is largely ignored,which has significant effects on the average network lifetime for those battery-limited wireless networks.In this paper,an opportunistic distributed best-relay node selection scheme is proposed for cooperative transmissions over underlay-paradigm based CR networks,and meanwhile guarantee that the primary link is provided with a minimum-rate for a certain percentage of time.The objective is to increase achievable data rate as well as prolong average network lifetime.Specifically,both the instantaneous channel state information (ICSI) and residual energy of candidate relay nodes are taken as weighted metric,and the relay node with the optimal weighted metric is selected as the best one.Simulation results are presented to demonstrate the effectiveness of the proposed scheme.