有向传感器网络量化跟踪技术研究

项目名称： 有向传感器网络量化跟踪技术研究

项目编号： No.61501448

项目类型： 青年科学基金项目

立项/批准年度： 2016

项目学科： 无线电电子学、电信技术

项目作者： 胡小青

作者单位： 中国科学院声学研究所

项目金额： 21万元

中文摘要： 有向传感器网络下目标跟踪问题逐渐引起人们的广泛关注。有向传感器不同于传统的全向传感器，其具有特定的视场角，不仅能提供探测到目标感知信号的模拟测量值，而且能给出目标的量化方向信息。现有的有向传感器网络跟踪算法通常单独地考虑目标探测方向和感知测量信息，没有在统一理论框架下有效地融合这两类信息，其很难保证获得到最小均方误差跟踪性能。本课题针对实际应用中有向传感器不确定性探测问题，建立有效的概率感知模型，并通过构建能有效融合量化的方向与模拟的测量信息的有向测量模型，在贝叶斯估计理论框架下开发新的数据融合方法及相应的低计算复杂度和高精度的近似方法，为有向传感器网络下目标量化跟踪提供解决途径。

中文关键词： 传感器网；目标跟踪；传感网测量；数据融合

英文摘要： The target tracking problem in directional sensor networks is attracting increasing attention. Unlike the traditional omnidirectional sensor, a directional sensor has a special angle of view. It can offer quantized direction information rather than just the analog measurement of the sensing signal with respect to the detected target. The existing tracking approaches in directional sensor networks always separately consider the direction and measurement information which have not been fused in the uniform framework; they hardly promise the tracking performance of minimum mean-square error. With the problem that the directional sensor has indeterminate detection ability, this project presents an effective probabilistic sensing model, and constructs a directional measurement model fusing the quantized direction and analog measurement information. A novel fusion method is proposed based on the Bayesian estimation theory, while the corresponding approximate algorithms of low computation complexity and high precision are also given to solve the quantized tracking problem in directional sensor networks.

英文关键词： sensor network;target tracking;sensor network measurement;data fusion