It is critical to maintain the communication links between important social pairs. However, maintaining the social links between faraway nodes in wireless networks is extremely difficult. Although multi-hop transmission can be used, if two nodes in the routing path are out of the wireless transmission range, a network partition is possible. To address this problem, we adopt a cooperative amplify-and-forward strategy, where nodes (relays) cooperate to improve the signal strength at the destination. We formulate and study two optimization problems for maintaining the required link throughput: Min-Energy and Min-Relay, where the goal of Min-Energy is to minimize the power consumption of the relays, and the goal of Min-Relay is to minimize the number of active relays. Since the Min-Energy problem is a non-convex problem, we solve it based on an approximation technique and prove that our solution is a feasible, in fact optimal solution. We formulate the Min-Relay problem as an integer programming problem, and propose a polynomialtime algorithm which can select the minimum number of relays to maintain the social link. Evaluation results show that Min-Relay can significantly reduce the number of active relays compared to Min-Energy, while achieving comparable power consumption.