In recent years, the workmanship of manufacturing sensors has achieved great success. The wide application of inexpensive hardware, such as complementary metal oxide semiconductor cameras and tiny microphones, has contributed to the development of WMSNs. WMSNs are composed of wirelessly connected devices that are able to collect information from the environment at any time. Most often, the types of information collected by WMSNs are video and audio streams, still images, and scalar sensor data. As with wireless sensor networks, the sensor devices in WMSNs are also strictly constrained in terms of memory, processing capability, storage capability, and especially energy. Although batteries can provide wireless sensors with energy, the energy is extremely limited. In practical applications, it is impossible for WMSNs to upload all of the multimedia data to users in remote regions. Hence, it is significant to develop energy-efficient techniques for data communication in WMSNs by in-network processing. In this article, a mechanism based on information fusion is proposed for reducing the volume of data being transferred. The mechanism is a trade-off between uploading the results of in-network data processing and uploading all of the raw data. Based on users' requirements, proper data will be uploaded, and the accuracy of querying will be as good as, or better than, uploading all of the raw data. Finally, the shortcomings and challenges of the mechanism are also described.
All Science Journal Classification (ASJC) codes