中文摘要：

近年来，无线传感器网络被广泛地应用到国民经济的各个领域．在传感应用中，轮廓查询作为多目标决策和数据挖掘的重要手段发挥着重要的作用．然而，由于轮廓查询需要消耗大量能量，因此在传感器网络中计算真实轮廓是非常不经济的．事实上，在大多数的应用中，近似轮廓已经可以反映数据的大致状态．通过对近似轮廓的语义研究，本文提出基于选择的基本近似轮廓算法（BAS）和基于映射的误差保证的近似轮廓算法（GAS）来高效地计算传感器网络中的近似轮廓．实验结果表明，这两种算法能有效地减少传感器网络中的通信代价，进而延长传感器网络的使用寿命．

英文摘要：

In recent years, wireless sensor networks have been widely used in many military and civil fields, especially in environmental monitoring. As an operator for multi-criteria decision making and data mining, skyline query is important in many sensing applications. However, due to the high computation cost of skyline queries, it is not energy efficient to conduct exact skyline over wireless sensor networks. In fact, in many applications, approximate skylines are enough for people to investigate the data and make decisions. Through exploiting the semantics of approximate skyline queries, two novel approaches, Basic Approximate Skyline algorithm （BAS） basing on selection and error-Guaranteed Approximate Skyline algorithm （GAS） basing on mapping, are proposed to evaluate the approximate skyline energy-efficiently in wireless sensor networks. Simulation results show that our approaches perform effectively on reducing communication cost and prolong the life-span of wireless sensor networks.