基于自回归积分滑动平均模型的无线传感网络通信传输信号延迟消除方法  

A Method for Eliminating Communication Transmission Signal Delay in Wireless Sensor Networks Based on Autoregressive Integral Moving Average Model

作　　者：崔蕾[1] 王同 CUI Lei;WANG Tong(Department of Information Engineering,Yantai Vocational College,Yantai Shandong 264670,China;School of Information Science and Engineeing,Shenyang Ligong University,Shenyang Liaoning 110159,China)

机构地区：[1]烟台职业学院信息工程系,山东烟台264670 [2]沈阳理工大学信息科学与工程学院,辽宁沈阳110159

出　　处：《传感技术学报》2025年第3期543-549,共7页Chinese Journal of Sensors and Actuators

基　　金：辽宁省自然科学基金项目(202135465550)。

摘　　要：为了解决受环境影响无线传感网络通信传输信号的延迟问题,提出了一种传输信号延迟消除的方法。将自回归积分滑动平均模型(ARIMA)和小波神经网络(WNN)相结合,进行通信传输信号延迟的组合预测。根据延迟预测结果设计传输信号延迟消除流程的步骤和约束条件,并以此构建无线传感网络通信传输的优化目标函数,引入免疫克隆蛙跳算法对目标函数进行求解,获取最优的传输方案。仿真分析表明,所提方法的延迟预测误差和端到端延迟误差低于0.01 s,能量消耗最大值为6.4 W,平均丢包率最大值为0.286%。上述结果证明了所提方法可以有效准确预测和消除无线传感网络通信传输信号延迟。In order to solve the problem of communication transmission signal delay in wireless sensor networks affected by the environ-ment,a method for eliminating transmission signal delay is proposed.by combining the autoregressive integral moving average model(ARI-MA)and wavelet neural network(WNN),integrated prediction of communication transmission signal delay is performed.The steps and constraints of the transmission signal delay elimination process are designed based on the delay prediction results,and an optimization ob-jective function is constructed for wireless sensor network communication transmission based on this.The immune clone leapfrog algorithm is introduced to solve the objective function and obtain the optimal transmission scheme.Results of simulation analysis shows that the de-lay prediction error and end-to-end delay error of the proposed method are less than 0.01 ms,the maximum energy consumption is 6.4 W,and the maximum average packet loss rate is 0.286%,demonstrating that the proposed method can effectively and accurately predict and eliminate communication transmission signal delays in wireless sensor networks.

关 键 词：无线传感网络 传输信号 延迟消除 自回归积分滑动平均模型 小波神经网络 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]