面向多重感知域的多模态传感器网络覆盖控制技术研究

The multiple sensor fields is one important application scenarios of IoT, Multimodal is the main feature of sensing layer of sensor network in multiple sensor fields. The qualities of coverage control strategies determine the performances and Qos of sensing of sensor networks. Coverage and connectivity models can significantly affect the coverage control for Multimodal Sensor Networks, However, the existing model can not be specialized to the strategies for polytype sensing modal in the multiple sensor field. Motivated by this, we propose and study the coverage control problem for Multimodal Sensor Networks in multiple sensor fields, taking full account of the polytype sensing modal and diverse communications needs. Two outcomes are expected, which are the new concepts, multimodal data fusion-centric coverage model and effective connectivity, which can be expected to characterize the influence in coverage and connectivity, by heterogeneity of sensing layer and communication layer for multiple sensor fields. Particularly, for the scenario of multiple sensor fields, the node scheduling in multiple sensor fields is focused on. New schemes based on cross-layer are expected to be proposed by combining Computational Intelligence, and Statistical Learning. The New schemes can effectively utilize the spatial temporal correlation and modal correlation of sensing data, and can fully meet the needs of the communication layer, so that the energy efficiency of multi-modal sensor nodes in the network is maximized. The achievements of the project will bridge the gap between the physical space and sensing space, and will be very useful for the further research and application of WSNs.

多重感知域是物联网中的重要应用场景，多模态是多重感知域内传感器网络感知层的主要特性。覆盖控制策略的优劣决定了感知的质量和性能，覆盖和连通模型是影响覆盖控制的重要因素，而现有模型并不适用于多重感知域。因此提出并研究多重感知域内多模态传感器网络覆盖控制的新问题，在充分考虑感知模态多样及其引发的通信需求多样的前提下，拟提出基于多模态数据融合的覆盖模型和有效连通的新概念，揭示多重感知域带来的感知层和通信层的异构性对覆盖连通的影响特性。提出并着重研究多重感知域内的节点调度问题。拟结合计算智能、统计学习等理论工具，制定适合多重感知域的跨层调度新方法。该方法能有效利用感知数据的时空相关性和模态相关性，并能充分满足通信层的需求，使多模态传感器网络中节点的能量效率最大化。该项目以多模态数据融合和有效连通覆盖模型为特色，架起物理空间与感知空间之间的桥梁，对传感器网络的研究和应用具有重要的学术价值。

多重感知域是物联网中的重要应用场景，多模态是多重感知域内传感器网络感知层的主要特性。覆盖控制策略的优劣决定了感知的质量和性能，覆盖和连通模型是影响覆盖控制的重要因素，而现有模型并不适用于多重感知域。因此提出并研究多重感知域内多模态传感器网络覆盖控制的新问题。在充分考虑感知模态多样及其引发的通信需求多样的前提下, 本项目研究了: 基于多模态数据融合的覆盖模型；多模态传感器网络有效连通的建模与分析；多模态传感器网络跨层调度的建模与分析。取得了下列理论成果：规则部署条件下基于数据融合的覆盖连通模型推导；基于数据融合的移动传感器网络部署策略设计；面向有效连通的感知数据安全分发策略设计；面向有效连通的频谱认知与接入策略设计；多模态感知中基于位置信息的时空感知策略设计；室内感知传输场景下可见光通信PAPR的抑制策略设计；基于下一跳状态估计能量收集传感器网络路由调度策略设计。项目的研究成果探索了物理空间与感知空间之间的桥梁，对传感器网络理论发展具有重要的学术价值，对物联网的大规模应用提供了一定的参考借鉴。