无线传感器网络中一种新的基于神经网络的自适应路由算法

A New and Adaptive Routing Algorithm Based on Neural Network in WSN

为了在无线传感器网络中建立合理有效的分簇结构,提高网络性能,最大限度地延长整个网络的寿命,提出了一种基于神经网络的自适应路由算法。提出的路由算法是在基站上采用神经网络的自适应学习来选择簇头,并根据最优权函数值来选择最短路径中的下一跳,同时添加一个新因子来选择网关节点进行簇头间的通信。实验结果表明提出协议的性能是EMHR的180%。该算法在节点中使用更少的计算和通信开销来构造和维护整个无线传感器网络,更加均衡网络负载,大大减少簇头节点成为网络瓶颈的概率,具有更强的网络生存能力和更长的生命周期。

For the sake of establishing reasonable and effective clustering structure of the network in WSN,improving the network performance and maximizing the life of the entire network,this paper presents an adaptive routing algorithm based on neural network clustering.It selects the cluster head by the adaptive learning of neural network on Base Station.Also,it chooses the next hop in the shortest path by the optimal value of the weighting function.Meanwhile it adds a new factor to select the gateway nodes for communication among cluster heads.The experimental results show that the performance of the proposed protocol is 108% of EMAR＇s.The former can use less computation and communication overhead to construct and maintain the entire network,balance network load,avoid the cluster head nodes becoming a network bottleneck and have a stronger network and longer life cycle.