基于深度学习模型的WSN路由协议算法
|
摘要
为降低无线传感网网络(WSN)的能量消耗,延长网络生存周期,提出一种基于深度学习模型的WSN路由协议算法。算法首先在汇聚节点完成训练并进行分簇,将训练好的参数传递给各簇节点,各簇节点对采集的数据进行特征分类、提取、再融合后传递给汇聚节点。在进行分簇时,为使簇头的分布更均匀,在估算最优簇头数的基础上,改进分簇方法,减少分簇次数,节省网络能量消耗。通过仿真实验表明,基于深度学习模型的WSN路由协议算法减少网络能量消耗,延长网络生命周期,更适合大规模远距离通信。
To reduce the energy consumption and prolong the lifetime of wireless sensor network(WSN), a WSN routing protocol algorithm based on deep learning model is proposed. Firstly, the algorithm completes training and clustering at the sink node, transfers the trained parameters to each cluster node, and then transfers the collected data to the sink node after feature classification, extraction and fusion. In order to make the distribution of cluster heads more uniform, the clustering method is improved on the basis of estimating the optimal number of cluster heads, which reduces the number of clusters and saves the energy consumption of the network. The simulation results show that the WSN routing protocol algorithm based on cascade automatic encoder reduces the network energy consumption, prolongs the network life cycle, and is more suitable for large-scale long-distance communication.
To reduce the energy consumption and prolong the lifetime of wireless sensor network(WSN), a WSN routing protocol algorithm based on deep learning model is proposed. Firstly, the algorithm completes training and clustering at the sink node, transfers the trained parameters to each cluster node, and then transfers the collected data to the sink node after feature classification, extraction and fusion. In order to make the distribution of cluster heads more uniform, the clustering method is improved on the basis of estimating the optimal number of cluster heads, which reduces the number of clusters and saves the energy consumption of the network. The simulation results show that the WSN routing protocol algorithm based on cascade automatic encoder reduces the network energy consumption, prolongs the network life cycle, and is more suitable for large-scale long-distance communication.
引文
[1]Gholipour M,Haghighat A T,Meybodi M R.Hop-by-hop traffic-aware routing tocongestion control in wireless sensor networks[J]EURASIP Journal on WirelessCommunications and Networking,2015(01):1-13
[2]杨庚,李森,陈正宇.传感器网络中面向隐私保护的高精确度数据融合算法[J].计算机学报,2013,36(01):189-200
[3]杨永健,刘帅.基于SOFM神经网络的无线传感器网络数据融合算[J].传感器技术学报,2013,26(12):1757-1763
[4]潘琢金,秦蓓,罗振,等.基于深度自编码器的WSN数据融合算法[J].计算机工程与设计,2018,39(05):1214-1218
[5]耿同同.基干LEACH协议的无线感器网络数据融合技本的研究[D].青岛:青岛科技大学,2017:5-10
[6]叶继华,王文,江爱文.一种基于LEACH的异构WSN能量均衡成簇协议[J].传感技术学报,2015(12):1853-1860
[7]侯彦军,谭国真.一种WSN分簇路由协议研究和实现[J].计算机科学,2015,42(05):160-164
[8]黄廷辉,伊凯,崔更申,等.基于非均匀分簇的无线传感器网络分层路由协议[J].计算机应用,2016(01):66-71
[9]Wang C,Jiang C,Liu Y,et al.Aggregation capacity of wireless sensor networks:Extended network case[J].Computers,IEEETransactions on,2014,63(06):1351-1364
[10]邱立达,刘天键,傅平.基于深度学习的无线传感器网络数据融合[J].计算机应用研究,2016,33(01):185-188
[11]王丽红,于光华,刘平.无线传感器网络LEACH算法的改进研究[J].齐齐哈尔大学学报:自然科学版,2018(03):2-4
[2]杨庚,李森,陈正宇.传感器网络中面向隐私保护的高精确度数据融合算法[J].计算机学报,2013,36(01):189-200
[3]杨永健,刘帅.基于SOFM神经网络的无线传感器网络数据融合算[J].传感器技术学报,2013,26(12):1757-1763
[4]潘琢金,秦蓓,罗振,等.基于深度自编码器的WSN数据融合算法[J].计算机工程与设计,2018,39(05):1214-1218
[5]耿同同.基干LEACH协议的无线感器网络数据融合技本的研究[D].青岛:青岛科技大学,2017:5-10
[6]叶继华,王文,江爱文.一种基于LEACH的异构WSN能量均衡成簇协议[J].传感技术学报,2015(12):1853-1860
[7]侯彦军,谭国真.一种WSN分簇路由协议研究和实现[J].计算机科学,2015,42(05):160-164
[8]黄廷辉,伊凯,崔更申,等.基于非均匀分簇的无线传感器网络分层路由协议[J].计算机应用,2016(01):66-71
[9]Wang C,Jiang C,Liu Y,et al.Aggregation capacity of wireless sensor networks:Extended network case[J].Computers,IEEETransactions on,2014,63(06):1351-1364
[10]邱立达,刘天键,傅平.基于深度学习的无线传感器网络数据融合[J].计算机应用研究,2016,33(01):185-188
[11]王丽红,于光华,刘平.无线传感器网络LEACH算法的改进研究[J].齐齐哈尔大学学报:自然科学版,2018(03):2-4