基于Bayes序贯估计的无线传感器网络数据融合算法

移动代理被认为是无线传感器网络中解决数据融合的有效方法，但代理访问节点的次序以及总数对算法有较大影响，为此该文提出一种基于Bayes序贯估计的移动代理数据融合算法.该算法通过构造特定数据结构的报文，在多跳环境中由Bayes序贯估计调整梯度向量，据此动态决定移动代理的访问路径，使移动代理有选择地在传感器节点之间移动，且在节点处由移动代理对数据进行融合，将多余的感知数据剔除，而不是把原始数据传输到Sink节点。理论分析和模拟实验表明，该算法有较小的能量消耗和传输延时。

A Data Fusion Algorithm Based on Bayes Sequential Estimation for Wireless Sensor Network

Mobile Agent(MA) is more suitable for wireless sensor networks than the C/S model in data fusion. In MA based data fusion, the order of nodes visited along the route by MA has a significant impact on the algorithm efficiency and life time of wireless sensor networks. This paper proposes a Mobile Agent Data Fusion (MADF) algorithm based on bayes sequential estimation for wireless sensor networks. By designing data packet and data table with specific structure, and considers MA in multihop environments and adopts gradient of Bayes sequential estimation to dispatch MA. MA accounts for performing data processing and making data aggregation decisions at nodes rather than bring data back to a central processor (sink), redundant sensory data will be eliminated. Theoretical analysis and experimental results show that the proposed scheme is able to provide less energy consumption and network delay compared to directed diffusion schemes.