基于群智感知的矿井灾后无线应急通信网络关键问题研究

The research on mine emergency communication networks is of great significance for the smooth progress of the rescue work. Due to the complicated environments and special constraints, it is difficult for the existing mine emergency communication systems to simultaneously meet various rescue requirements, such as energy, efficiency, and accuracy. Therefore, the mine emergency communication networks need to be further explored and researched. The basic theories and methods for the wireless emergency networks are studied based on the crowd sensing technology. The outline of this study is listed as follows: (1) Aiming at the energy constraints of mobile terminals in the mine disaster rescue, the optimization of the sensing abilities is studied from the aspect of sensing layer, and a collaborative assignment strategy for the sensing tasks and nodes is proposed. (2) In consideration of the time efficiency requirements in the mine disaster rescue, the quick asynchronous neighbor discovery strategy is researched from the aspect of network layer. Then, the efficient opportunistic transmission is explored. (3) In order to meet the demands of the data accuracy, the observation model is constructed for the mine emergency communication system from the aspect of application layer, where the complicated environments and the opportunistic routing are considered simultaneously. The intelligent accurate estimation strategy is proposed based on the model constructed. Moreover, the stability of the proposed estimation strategy is discussed. The project plans to obtain series of theoretical results. The results obtained will provide a new idea for developing the mine emergency networks as well as has a high application value in the field of mine disaster rescue.

矿井应急通信网络对于保障矿难救援工作的顺利开展具有重要意义。矿难灾后环境复杂恶劣、制约性强，现有应急通信系统难以满足井下救援对能量、效率以及精度等方面的要求，有待于对其进一步深入探索研究。本项目基于群智感知研究矿井灾后无线应急通信网络基础理论与方法，主要内容包括：（1）针对矿井灾后救援终端能量受限问题，从感知层面研究感知能力优化组合方法，探索感知任务与感知节点协作分配方案。（2）考虑矿井灾后救援的高时效性要求，从网络层面研究快速异步邻居发现策略，进一步探讨高效机会传输机制。（3）基于矿井灾后救援对数据准确性的要求，从应用层面构建考虑井下环境特点和机会路由的系统观测模型，据此设计智能精确数据估计策略并分析其稳定性能。本项目的研究将得到系列性理论成果，为矿井灾后应急通信网络的设计提供新思路，同时对矿难救援也具有重要实际应用价值。

煤炭是目前我国的主体和支柱能源。随着开采规模的日益扩大，矿难时有发生。应急通信网络对于保障矿难救援工作的顺利开展具有重要意义。本项目面向矿井灾后应急救援国家重大需求，深入研究了基于群智感知的灾后应急网络重构理论与方法，主要研究内容包括：（1）矿井灾后应急网络感知优化方法。提出了基于LoRa的长距离井下应急救援手势感知算法，设计了基于低成本无源光学标签的矿难救援机器人位置感知方法，构建了融合多种感知手段的协作感知框架，实现了群体化、高效化感知。（2）矿井灾后应急网络数据传输机制。研究了跨网通信优化技术，给出了具有节能和加速两种模式的事件驱动邻居发现策略，兼顾能耗与发现时延，实现了通信多元化、转发即时化，为矿井灾后救援应用提供网络支撑。（3）矿井灾后应急网络数据估计方法。提出了矿井弱连通环境下的自适应鲁棒时钟同步算法，设计了一种融合超宽带（UWB）和惯性测量单元（IMU）的井下无人机按需精确跟踪方法，在保证定位精度的前提下有效降低了能耗，为矿井灾后应急救援提供精准信息。经过三年研究，本项目取得了系列性研究成果，对于提高矿井灾后救援效率具有较高的理论价值和实践意义。