Wireless sensor networks (WSN) are ad-hoc wireless networks used in many domains involving the analysis of various properties. Along with the growth in the number of deployments of WSNs, there has been an increase in the size of the networks as well. These systems tend to generate a tremendous amount of data that require further processing and analysis. Transmission of data requires a huge amount of energy considering that the components are powered by tiny batteries with a short life cycle. Consequently, energy efficiency is a dominant factor in the performance measure of WSNs. One major contributor to the overall energy consumption of a WSN is overhearing, a condition, in which the same data is picked up and transmitted by multiple nodes. Reducing overhearing results in a marked decrement in energy usage. This paper proposes to use semantic technology for the observation and collection of sensor data. With the use of an ontology, sensor nodes can be remarkably interoperable and configurable in the receipt and transmission of semantic data. Using this interoperability, we introduce asynchronous semantic preamble listening to avoid overhearing in a semantic sensor network. Performance comparisons to LPL by real experiments show stark improvements in energy consumption.

无线传感器网络（WSN）是在许多领域中使用的自组织无线网络，涉及对各种属性的分析。随着WSN部署数量的增加，网络规模也随之增加。这些系统倾向于生成大量需要进一步处理和分析的数据。考虑到组件由寿命短的微型电池供电，数据传输需要大量能量。因此，能源效率是无线传感器网络性能衡量的主要因素。WSN总体能耗的一个主要贡献是偷听，这种情况是多个节点收集并传输相同的数据。减少偷听行为会导致能源使用量显着减少。本文提出使用语义技术来观察和收集传感器数据。通过使用本体，传感器节点在语义数据的接收和传输中可以显着地互操作和可配置。利用这种互操作性，我们引入了异步语义前导侦听，以避免在语义传感器网络中发生偷听行为。实际实验与LPL的性能比较表明，能耗显着提高。