车辆异构网络中基于决策树的稳健垂直切换算法

发布时间：2018-05-16 21:30

本文选题：车辆异构网络 + 垂直切换　；参考：《电子与信息学报》2017年07期

【摘要】：在带有噪声干扰的车辆异构网络中,针对当前基于决策树的垂直切换算法存在切换精准性不高的问题,该文详细分析并给出当前算法决策过程中存在的错误判决概率,提出一种基于决策树的稳健垂直切换算法。首先,采用卡尔曼滤波算法,根据网络属性的预测值和当前测量值,并结合它们各自的噪声偏差,做出更准确的网络属性估计。其次,针对少量网络属性值出现在判决门限附近的情况,提出概率阈值区间法,通过二次检测提高算法判决的准确性。仿真结果表明,所提算法提高了切换判决精准性和网络总吞吐量,降低了乒乓效应和切换失败率,并得到了与传统算法同阶的时间复杂度性能结果。
[Abstract]:In the vehicle heterogeneous network with noise interference, aiming at the problem that the current vertical handoff algorithm based on decision tree is not accurate enough, this paper analyzes and gives the error decision probability in the decision-making process of the current algorithm in detail. A robust vertical handoff algorithm based on decision tree is proposed. Firstly, Kalman filtering algorithm is used to estimate the network attributes more accurately according to the network attribute prediction value and the current measurement value, and combining their respective noise deviations. Secondly, a probability threshold interval method is proposed to improve the accuracy of the algorithm. Simulation results show that the proposed algorithm improves the precision of handoff decision and the total throughput of the network, reduces the ping-pong effect and the handoff failure rate, and obtains the performance results of the same order of time complexity as the traditional algorithm.
【作者单位】：重庆邮电大学重庆市计算机网络与通信技术重点实验室;重庆邮电大学计算机科学与技术学院;
【基金】：国家自然科学基金(61601070,61271259,61301123,61471076) 重庆市基础与前沿研计(cstc2016jcyj A0455,cstc2015jcyj A40047) 重庆邮电大学博士启动基金(A2014-10,A2015-16)~~
【分类号】：TN713;TN915.0

【相似文献】