The limitation on the battery lifetime has been a critical issue for the advancement of mobile computing. Different types of power-saving techniques have been proposed in various fields. In order to provide feasible energy-conserving mechanism for the Mobile Subscriber Stations (MSSs), three power-saving types have been proposed for the IEEE 802.16e broadband wireless networks. However, these power-saving types are primarily targeting for the cases with a single connection between the Base Station (BS) and the MSS. With the existence of multiple connections, the power efficiency obtained by adopting the conventional scheduling algorithm can be severely degraded. In this paper, a Maximal Power-Conserving (MPC) scheduling algorithm is proposed to consider the aggregated effect from the multiple connections to the power efficiency. Moreover, the Quality-of-Service (QoS) constraints from both the downlink and the uplink traffic are employed in the design of the MPC algorithm in order to facilitate the corresponding MSS to fulfill its QoS requirements in both directions. Numerical results show that the proposed MPC scheduling algorithm outperforms the conventional 802.16e power-saving mechanism, especially under the multi-connection scenarios.