通信约束下的水声信号分布式检测融合算法研究

针对水声传感器网络中大延迟、低可靠通信约束下的水声信号分布式检测问题，提出了一种基于时间窗口的自适应融合算法．传感器节点依据声纳接收机的特性计算局部判决并发送给融合中心节点．融合中心节点在时间窗口内，基于已收到的局部判决在线自适应地调整融合规则，从而由最优融合算法得到最终判决．通过仿真，讨论了时间窗口的选择问题以及融合算法的性能．仿真结果表明，新算法具有很高的实用性，能够在动态变化的水声通信条件下保证整个系统高效运行．

A Fusion Algorithm for Distributed Detection of Underwater Acoustic Signal under Communication Constraints

An adaptive fusion algorithm based on time window is proposed for distributed detection of underwater acoustic signal under communication constraints, such as large propagation delay and low reliability in underwater acoustic sensor networks. Each sensor node makes a local decision according to the characteristics of sonar receiver and sends it to the fusion center node. Within the time window, this node adjusts the decision rule on line based on the received local decisions. Thus, a final decision is made according to the optimal fusion algorithm. The selection of time window, as well as the algorithm performance is further discussed by simulation. Simulation results show that the new algorithm is highly prartieal, and systems with the proposed algorithm can function effectively in dynamically variational underwater acoustic communication.