LEACH—DRT算法中基于神经网络的数据预测算法

为了保证无线传感器网络中数据的完整性，针对基于LEACH路由协议的动态轮时间算法存在的问题，提出一种基于人工神经网络的数据预测算法。该动态轮时间算法中，部分簇因调整后的轮时间不足以完成数据的采集而丢失数据。数据预测算法结合传感器节点数据具有时空相关性的特点，将时空延迟算子引入神经网络模型，并通过建立的神经网络模型对数据进行预测。仿真时采用伯克利英特尔实验室的传感器数据，通过Mafl软件对模型进行测试并分析仿真结果。实验结果表明：该算法对连续多个数据的预测效果理想，预测误差始终保持在较低水平。

A Data Prediction Algorithm Based on Artificial Neural Network in LEACH-DRT Algorithm

To address at the problems that existed in dynamic roundtime algorithm based on low energy adaptive clustering hierarchy protocol （LEACHDRT）, the data predictive algorithm based on artificial neural network was proposed to guarantee the in tegrity of data in wireless sensor networks. In LEACHDRT algorithm, part of clusters could not finish data acquisition because of lack of round time, and then it leads to data loss. In terms of to characteristics of sensor nodes with spatialtemporal correlativity, al gorithm was introduced in spatialtemporal delay operators into neural networks model and data were predicted with the model. The Berkeley Intel lab＇s sensor data streams were adopted in simulation as empirical data, the simulation results were predicted with model by Matlab software. Simulation result proved that algorithm achieved ideal effect in predicting consecutive data, and predic tion error stayed at a lower level.