基于蚁群神经网络的两级信息融合算法

A Two Level Information Fusion Algorithm Based on Ant Colony Neural Network

为了保证地下车库空气质量的同时尽量降低系统能耗,针对地库环境监测系统,提出了一种基于蚁群神经网络的两级数据融合算法TLIFA-ACOBP,该算法将分簇结构与神经网络模型有效结合,设计了一个基于分簇的无线传感器网络两级数据融合模型.首先运用蚁群优化(ACO)算法对BP神经网络的权值进行优化,并将优化后的蚁群神经网络用于无线传感器网络的信息融合.通过对簇成员节点采集到的原始数据进行两级融合处理,只将代表原始数据的少量特征值发送给汇聚节点,大幅度减少节点数据通信量,提高了数据传输效率,同时降低了系统能耗.

In view of the garage environment monitoring system,in order to ensure the air quality of garages and reduce the energy consumption to the minimum,this paper presents a two-level data fusion algorithm,named TLIFA-ACOBP,based on the ant colony and neural network.The algorithm combines the cluster structure with the neural network model effectively and designs a two-level data fusion model based on the cluster structure of wireless sensor network（WSN）.First,the ant colony optimization（ACO）algorithm is used to optimize the weights of BP neural network.Then the optimized ant colony neural network is applied to the information fusion of WSN.By the two level fusion processing,the original collected data are transformed into a few representative characteristic values which are sent to the sink node subsequently.Thus the TLIFA-ACOBP reduces the amount of communication workload substantially,improves the efficiency of data transmission,and decreases the energy consumption of the system.