基于时-空联合压缩的多媒体传感网数据收集技术研究

As the senior type of wireless sensor networks and basic data collection network for internet of things, wireless multimedia sensor networks is an actively and hot research area. Different from energy focalize consumption due to scalar data transmission in traditional wireless sensor networks, energy consumption is almost evenly distributed according to multimedia data sensing, processing and transmission in wireless multimedia sensor networks. Thus data collection methods of traditional networks can not be applied directly. This project investigates on video data collection for wireless multimedia sensor networks based on spatial-time joint compression, aims to design the perceptual model of wireless multimedia sensor nodes and coverage control mechanism, research the in-network nodes' cooperation image compression method based on temporal-spatial correlation, explore the QoS distinguished image transmission mechanism for wireless multimedia sensor networks. On the basis of simulations and actual deployments, a prototype system will be developed for image collection and transmission in ecological environment monitoring based on wireless multimedia sensor networks. Results of our project will provide theoretical basis and technical support for the development and popularization of wireless multimedia sensor network in new application scenarios such as environmental monitoring, intelligent monitoring, etc.

作为无线传感器网络高级形式和物联网数据感知基础设施的无线多媒体传感器网络是当前极具活力和影响的研究领域。不同于传统传感器网络能量消耗主要集中于标量数据的无线传输，无线多媒体传感器网络能耗均匀分布在视频图像数据的感知、处理和传输，亟需研究高效节能的无线多媒体传感器网络视频图像数据收集方法。本项目将重点研究无线多媒体传感器网络节点感知模型及覆盖控制机制、基于时空关联的节点协作网内信息压缩处理方法以及面向区分服务质量保证的无线多媒体传感器网络图像传输技术等；在此基础上构建一个面向生态环境监测图像收集的无线多媒体传感器网络应用示范，验证上述关键理论和算法的可行性和有效性。项目研究成果可为无线多媒体传感器网络在生态环境监测、智能监控等典型应用提供理论依据和技术支撑。

本项目重点围绕面向三维场景的无线多媒体传感网感知模型及其覆盖控制机制、无线传感网定位理论与方法、基于空间冗余消除的无线多媒体传感器网络数据压缩方法、高效的无线传感网数据收集方法以及普适环境下的新型场景感知技术等关键问题开展研究，取得较好的研究进展，相关成果在INFOCOM、ToN、JSAC、TVT、TECS、TETC等重要会议和期刊上发表，并获得多项国家发明专利授权和科技奖励。