传感器网络中基于移动代理的数据融合方法研究

The Mobile Agent-Based Data Fusion in Sensor Network

在传感器网络中采用移动代理来进行数据融合是一个新颖的思路,它与传统的数据融合方法相比拥有诸多优势,有必要为数据融合设计一种基于移动代理的计算模型,使得移动代理在传感节点间迁移的同时能够进行有效的数据融合。我们对基于移动代理的数据融合方法进行了深入的探索,设计了基于移动代理的数据融合框架,提出了一种与移动代理路由紧密结合的按分辨率并行量化交叠的数据融合算法——PQOR,并将其成功地运用到目标分类识别的应用场景中。仿真结果表明与传统的数据融合算法相比,PQOR能够以较小的代价达到应用的要求,其优势随着网络节点规模的增长更为明显。

Mobile agent-based method for data fusion has many advantages over the traditional methods for data fusion in sensor network. It is quite necessary to develop a mobile agent based computing model for data fusion, where a mobile agent selectively visits the sensors and incrementally fuses the appropriate measurement data. We design a mobile agent-based framework for data fusion in wireless sensor network. Also we propose an algorithm of parallel quantifying overlapping on resolution named PQOR, which is successfully applied for data fusion in the scenario of target classification. The simulation results show that PQOR has a superior overall performance over the existing algorithms and its advantage becomes more evident as the node size increases.