异构无线网络中基于人工神经网络的自适应垂直切换算法

针对当前基于人工神经网络的垂直切换算法(ANN-VHO)，存在业务自适应性差和计算复杂度高的问题，该文提出一种基于人工神经网络的自适应垂直切换算法。首先，根据终端获取到的接收信号强度(RSS)，采用阈值判断的方法，遴选出候选网络集；其次，根据该文划分的不同业务类型，对参数进行自适应选择和归一化；再次，把选择的参数输入人工神经网络，判决出候选网络集中最佳的接入网络。最后，实验结果表明，该算法能根据用户的业务类型合理地选择切换网络，降低切换阻塞率，同时降低算法的时间复杂度。

An Adaptive Vertical Handover Algorithm Based on Artificial Neural Network in Heterogeneous Wireless Networks

Current research on Vertical HandOver algorithm based on Artificial Neural Network (ANN-VHO) has a poor service adaptability and high computational complexity. Considering this problem, an adaptive vertical handover algorithm based on artificial neural network is proposed. Firstly, according to the Received Signal Strength (RSS) obtained by the terminal, a method of thresholding is used to select a candidate network set. Secondly, in terms of the different types of services classified in this paper, the parameters are normalized and adaptively selected; Thirdly, the selected parameters are input into the artificial neural network to choose the best access network from the candidate network. Finally, the experimental results show that the algorithm can reasonably select the handover network according to the user's service type, reduce the handover blocking rate and lower the time complexity of the algorithm.