One of the main issues in wireless sensor networks (WSNs) is energy conservation due to limited power of sensor node battery. To save energy, WSNs operate in duty-cycling work mode, when the sensor nodes alternate between active and dormant states. In this paper, we propose an active node set selection algorithm (ANSS), based on preference aggregation. The aim of the algorithm is to reduce the number of active sensor nodes, thus decreasing energy consumption while keeping required measurement accuracy level. Nodes, included in the active node set, are selected on the base of such parameters as energy consumption, distance to a cluster head, and multisensor general accuracy.